geo report - The University of Hong Kong

Transcription

THE UNIVERSITY OF HONG KONG
LIBRARIES
This book was received
in accordance with the Books
Registration Ordinance
Section 4
INCIDENTS IN 1997
(VOLUME 1)
GEO REPORT No. 79
Halcrow Asia Partnership Ltd.
GRDINAIICE
Chapter 142
No,: HK
lJ
^
!/•>
RI
- 2 -
© The Government of the Hong Kong Special Administrative Region
First published, November 1998
Prepared by:
Geotechnicat Engineering Office,
Civil Engineering Department,
Civil Engineering Building,
101 Princess Margaret Road,
Homantin, Kowloon,
Hong Kong.
r
,,„_ ^r.^,-
LiiJLf!Ei^^
«p
, ?*<
fr
,
This publication is available from :
Government Publications Centre,
Ground Floor, Low Block,
Queensway Government Offices,
66 Queensway,
Hong Kong.
Overseas orders should be placed with:
Publications Sales Office,
Information Services Department,
28th Floor, Siu On Centre,
188 Lockhart Road, Wan Chai,
Hong Kong.
Price in Hong Kong : HK$192
Price overseas : US$30 (including surface postage)
An additional bank charge of HK$50 or US$6.50 is required per cheque made in currencies
other than Hong Kong dollars.
Cheques, bank drafts or money orders must be made payable to
The Government of the Hong Kong Special Administrative Region
- 3-
PREFACE
In keeping with our policy of releasing information which
may be of general interest to the geotechnical profession and the
public, we make available selected internal reports in a series of
publications termed the GEO Report series. A charge is made to
cover the cost of printing.
The Geotechnical Engineering Office also publishes
guidance documents as GEO Publications. These publications
and the GEO Reports may be obtained from the Government's
Information Services Department. Information on how to
purchase these documents is given on the last page of this report.
R.K.S. Chan
Principal Government Geotechnical Engineer
November 1998
- 4-
EXPLANATORY NOTE
This GEO Report consists of four Landslide Study Reports on the investigation of
selected slope failures that occurred in 1997. The investigations were carried out by Halcrow
Asia Partnership Ltd (HAP) for the Geotechnical Engineering Office as part of the 1997
Landslip Investigation Consultancy.
The LI Consultancies aim to achieve the following objectives through the review and
study of landslides:
(a) establishment of
methodology,
an
improved
slope
assessment
(b) identification of slopes requiring follow-up action, and
(c) recommendation of improvement to the Government's
slope safety system and current geotechnical engineering
practice in Hong Kong.
The Landslide Study Reports prepared by HAP are presented in four sections in this
Report. Their titles are as follows :
Section
Title
Page No.
1
Detailed Study of the Landslide at University KowloonCanton Railway Corporation Station on 2 July 1997
5
2
Detailed Study of the Landslide near Kowloon-Canton
Railway Corporation Fo Tan Station on 2 July 1997
34
3
Detailed Study of the Flooding Incident at Mei Chung
Court, Shatin on 2 July 1997
66
4
Detailed Study of the Flooding Incident at Tuen Mun Road
Yau Kom Tau on 2 and 3 July 1997
103
The Landslip Investigation Division of the Geotechnical Engineering Office worked
closely with the LI Consultants and provided technical input and assistance to the landslide
studies.
- 5-
SECTION 1 :
KOWLOON-CANTON
STATION ON 2 JULY 1997
Halcrow Asia Partnership Ltd
This report was originally produced in March 1998
as GEO Landslide Study Report No. LSR1/98
- 6 -
FOREWORD
This report presents the findings of a detailed study of a landslide
(GEO Incident No. ME97/7/29) which occurred on the 2 My 1997 on
a slope south of the Kowloon-Canton Railway Corporation (KCRC)
University Station. Debris from the landslide obstructed the northbound KCRC rail track and caused significant disruption to rail
services throughout the morning. No fatalities or injuries were
reported.
The key objectives of the detailed study were to document the
facts about the landslide, present relevant background information and
establish the probable causes of the landslide. The scope of the study
was generally limited to site reconnaissance, desk study and analysis.
Recommendations for follow-up actions are reported separately.
The report was prepared as part of the 1997 Landslip
Investigation Consultancy (LIC), for the Geotechnical Engineering
Office (GEO), Civil Engineering Department (CED), under
Agreement No. CE 68/96. This is one of a series of reports produced
during the consultancy by Halcrow Asia Partnership Ltd (HAP). The
report was written by Mr M Riley and reviewed by Dr R Moore and
Mr H Siddle. The assistance of the GEO in the preparation of the
report is gratefully acknowledged.
G. Daughton
Project Director/Halcrow Asia Partnership Ltd.
- 7 -
CONTENTS
Page
No.
Title Page
5
FOREWORD
6
CONTENTS
7
1.
INTRODUCTION
8
2.
THE SITE
8
2.1
Site Description
8
2.2
Site History
9
2.3
Previous Studies
10
2.4
Subsurface Conditions
11
3.
THE LANDSLIDE
12
3.1
Time of Failure
12
3.2
Description of the Landslide
12
4.
RAINFALL
13
5.
PROBABLE CAUSES OF FAILURE
14
6.
CONCLUSIONS
15
7.
REFERENCES
15
LIST OF TABLES
17
LIST OF FIGURES
19
LIST OF PLATES
29
- 8-
1.
INTRODUCTION
On the morning of 2 July 1997, a landslide (GEO Incident No. ME97/7/29) occurred on
a slope south of the Kowloon-Canton Railway Corporation (KCRC) University Station
(Figure 1). Debris from the landslide blocked the north-bound KCRC rail track and caused
significant disruption to rail services throughout the morning. No fatalities or injuries were
reported.
Following the landslide, Halcrow Asia Partnership Ltd (the 1997 Landslip
Investigation Consultants) carried out a detailed study of the failure for the Geotechnical
Engineering Office (GEO), Civil Engineering Department (CED), under Agreement
No. CE 68/96. This is one of series of reports produced during the consultancy by Halcrow
Asia Partnership Ltd (HAP).
The key objectives of the study were to document the facts about the landslide, present
relevant background information and establish the probable causes of failure. The scope of
the study was limited to site reconnaissance, desk study and analysis. Recommendations for
follow-up actions are reported separately.
This report presents the findings of the detailed study which comprised the following
key tasks:
(a) a review of relevant documents relating to the history of the
site,
(b) analysis of rainfall records,
(c) interviews with KCRC staff involved in the incident,
(d) detailed site observations and measurements at the landslide,
and
(e) diagnosis of the probable causes of failure.
2.
THE SITE
2.1
Site Description
The location of the landslide is shown in Figure 2. The failure occurred at the crest of
registered cut slope No. 7NE-C/C106 and extended upslope into the sloping ground below Tai
Po Road. The total height of the slope, from the toe of the cut to Tai Po Road, is about 28 m.
The KCRC north-bound rail track is approximately 3 m from the toe of the cut slope section.
At the location of the landslide the dominant slope angle is about 40°. The inclination
of the lowest 2 m of the slope is about 60°. There is no discernible difference in slope angle
across the apparent boundary between the cut slope and the ground above its crest shown on
the topographic map. The true extent of cutting is uncertain.
- 9-
Prior to failure the lower 6 m section of the cut slope had been shotcreted, but elsewhere
the slope was densely vegetated (Figure 2 and Plate 1).
A 2.5 m high concrete retaining structure, approximately 3 m long and inclined at about
55°, is present at the toe of the slope immediately south of the failed section (Figure 2).
No surface drainage measures are present on the slope, but a covered U-charmel
associated with the railway is located near the toe. Tai Po Road is located at the crest of the
slope about 8 m beyond the crest of the landslide. The road falls to the south at about 5° which
directs surface water flow to a drainage system located at the western (inside) kerb of the road
and away from the failure location. A 100 mm diameter metal pipe, thought to be a water
carrying installation, was observed at the ground surface near the crest of the slope. Subsurface,
water-carrying services are also present In the eastern pavement of Tai Po Road associated with
a fire hydrant located about 10m north of the landslide site.
The landslide principally occurred on land within the KCRC boundary but its western
part also affected unallocated Government land (Figure 2). Sha Tin Town Lot (STTL) No. 437
is located near to the northern part of the landslide but was not directly affected by the incident.
2.2
Site History
KCRC records indicate that many of the slopes along the railway corridor remained
"virtually untouched since their construction in 1910" (Ove Arup & Partners, 1987a) until the
mid 1980's. At that time, a programme for upgrading the railway, involving double-tracking
and realignment of the route was completed, and involved modification to some slopes. KCRC
have no records of any modifications made to cut slope No. 7NE-C/C106 during these works.
The cut slope can be seen on the earliest available aerial photographs of the area, taken
in 1963. Features which appear to be shallow failures were evident in the central and northern
parts of the cut slope in photographs taken In 1964 and 1980 respectively (Figure 3). No
records relating to these incidents have been located.
The cut slope has been densely vegetated for much of its history though the lower part
of the slope was shotcreted in 1991.
Following the recent landslide, loose debris was removed and the scar was shotcreted by
KCRC at the recommendation of the GEO. Weepholes were installed at regular intervals to
enable drainage of water from the slope.
GEO recommended to Buildings Department (BD) that a Dangerous Hillside Order
(DHO) be served on KCRC with respect to the failed slope. The DHO, however, had not been
served at the time of preparation of this report.
Following this and other recent failures, KCRC have instigated a geotechnical review of
slopes within their boundaries and adjacent to the railway tracks.
- 10 -
2.3
Previous Studies
Cut slope No. 7NE-C/C106 was registered in the 1977/78 Catalogue of Slopes and a
field sheet was prepared by the consultants engaged by Government to prepare the Catalogue
following an inspection of the slope in April 1978 (Binnie & Partners, 1978). No visible
evidence of previous instability or distress was identified at that time.
In 1984, KCRC began a comprehensive geotechnical investigation of all existing slopes
adjacent to the newly upgraded rail tracks. The project was divided into three stages, namely:
(a) Stage 1: Preliminary inspection and cataloguing of cut slopes
and embankments between Hung Horn and Lo Wu to identify
slopes requiring further study. Inspection data were then
considered together with the results of a quantitative stability
assessment of all cut slopes carried out using assumed soil
parameters. Embankments were considered to present a
lower risk to the railway than cut slopes and were therefore
excluded from further study. The findings were presented in
a Preliminary Report in 1984 (Ove Arup & Partners, 1984).
(b) Stage 2: Geotechnical investigation of 24 slopes identified
during Stage 1 as posing a threat to the railway or to
structures and private property above the slopes. The
assessments were based on a programme of site specific
ground investigation and laboratory testing. The findings
were presented in a report dated April 1987 (Ove Arup &
Partners, 1987b).
(c) Stage 3: Detailed design of remedial works for "fourteen
slopes identified as not meeting current standards" (Ove Arup
& Partners, 1987a) and the letting and supervision of the
construction contracts.
Cut slope No. 7NE-C/C106 was selected for further study as part of the Stage 2
investigation programme which began in March 1986.
The Stage 2 investigation determined that remedial works were required to improve
the factor of safety (FOS) of the slope against potential deep failure to the desired standard
(FOS = 1.2). The proposals recommended installation of doweled ground beams over a
central 45 m length, and minor trimming and surfacing of the lower slopes at either end of the
cutting (Ove Arup & Partners, 1987b). KCRC have confirmed that "works, including
removal of loose materials, installation of dowel bars and (application of) shotcrete to the
slope surface to the height of 10 m from the toe" were completed in 1991, but noted that the
recent landslide originated at the "upper part of the slope where no improvement work has
been carried out" (KCRC, 1997a). KCRC have no records to indicate that the design of the
slope improvement works was checked and approved by the GEO prior to construction. A
- 11 review of GEO records also found no evidence that the design had been submitted to GEO for
checking.
Cut slope No. 7NE-C/C106 was considered for Stage 1 Study by the Geotechnical
Control Office (GCO) in 1988, but was not selected because the "slope (had) already been
investigated to at least Stage 1 Study level by KCRC's consultant" (GCO, 1988).
KCRC carry out annual inspections of slopes adjacent to the railway tracks (KCRC,
1997b). The cut slope is registered as Slope No. U14.3 (7.10) in the KCRC system and was last
inspected before the landslide on 3 May 1997. The inspection report (KCRC, 1997c) concluded
that the slope was in generally good condition, and recommended only minor maintenance
works including removal of blockages noted in weepholes. It is not clear whether the ground
above the crest of the cut slope was included in the inspections.
In mid-1992, the GEO initiated a consultancy agreement entitled 'Systematic Inspection
of Features in the Territory' (SIFT) to update information on existing registered slopes based on
studies of aerial photographs. Cut slope No, 7NE-C/C106 was identified as a Class Cl feature
by SIFT in 1994. According to SIFTs classification system, Class Cl features are "cut features
considered to meet GEO criteria for slope registration" and "have been formed or substantially
modified before 30.6.1978 or have been illegally formed after 30.6.1978" (King, 1994).
In 1994, the GEO commenced a consultancy agreement entitled 'Systematic
Identification and Registration of Slopes in the Territory (SIRST) to systematically update the
1977/78 Catalogue of Slopes and prepare the New Catalogue of Slopes. The SIRST field sheet,
prepared for cut slope No. 7NE-C/C106 in July 1995 indicates that Stage 1 Study was not
required as the feature had previously been "checked by GEO".
2.4
Sheet 7 of the Hong Kong Geological Survey 1:20 000 scale map (GCO, 1986) and the
1987 Geotechnical Area Studies Programme Engineering Geology Map for the Central New
Territories (GCO, 1987), indicate that the site is underlain by medium-grained granite.
Ground investigation data obtained for the KCRC Stage 2 investigation of the cut slope
in 1986 (Bachy Soletanche Group, 1986) included two boreholes and three surface strips at the
locations shown in Figure 3.
Borehole BH18, sited upslope of the landslide (Figure 3), encountered slightly
decomposed medium- to coarse-grained granite at a depth 9.3 m, overlain essentially by
completely decomposed granite (CDG). Surface strip CS3, located nearest to the landslide site,
exposed completely decomposed granite over most of its length, though the lower 7.3 in were
recorded as highly to moderately decomposed granite.
Borehole BH17, located about 50 m south of the landslide (Figure 3) proved slightly
decomposed, coarse-grained granite at a depth of 10.8 m overlain by a variable sequence
(6.8 m thick) of moderately to completely decomposed granite. The borehole also recorded a
4 m thickness of "loose greyish and yellowish brown silty fine to coarse sand" which was
- 12 -
classified as fill (Bachy Soletanche Group, 1986). Surface strips SI and S2 also identified
"very loose" fill materials which extended 6.1 m and 15.5 m from the top of the exposed area
respectively. The fill was recorded on the surface strip logs to have been placed by "dumping"
(Bachy Soletanche Group, 1986).
The borehole and surface strip logs show the presence of a loose fill body at the crest of
the southern part of the slope away from the location of the recent landslide site. The possible
extent of the fill body, based on interpretation of KCRC's ground investigation data is shown in
plan in Figure 3 and in section on Figure 4.
Piezometers were installed in BH17 at 7 m depth in June 1986 and at 9 m depth in
BH18 in July 1986. These were reported as dry in the 1987 Stage 2 geotechnical assessment
report (Ove Arup & Partners, 1987b).
3.
THE LANDSLIDE
3.1
Time of Failure
The landslide was first observed by the driver of a south-bound train that was waiting at
University Station early on the morning of 2 July 1997. The train was scheduled to arrive at the
station at 06:00 hours, and the driver noted a "big noise at about 8 minutes after arrival"
(KCRC, 1998). The driver then left the train and saw the landslide debris obstructing the northbound tracks. The landslide most probably occurred therefore at about 06:08 hours on the
morning of 2 July 1997. A landslide warning was issued at 06:25 hours on 2 July 1997.
3.2
Observations made during inspections of the landslide are annotated in Figure 5 and a
cross-section through the failed section is given in Figure 6. Photographs of the landslide are
shown in Plates 1 to 4.
The landslide occurred 8 m below Tai Po Road, producing a rapture surface about
9.2 m long and up to 7 m wide. The volume of failure was estimated at about 30 m3. The
landslide involved both the cut slope and the ground over its crest. The failure was shallow,
reaching a maximum depth of about 1 m.
The failure scar exposed between 0.6 m and 1 m of granitic residual soil overlying
completely decomposed fine- to medium-grained granite (Figure 6). A planar surface inclined
at 55°, possibly a relict discontinuity, was exposed in the completely decomposed granite at the
upper part of the failure scar.
No seepage was observed at the failure scar during inspections of the landslide on
4 July and 22 July 1997.
The debris spread out as it ran down the cut slope forming a debris fan about 12m wide
at the toe of the cut slope. The majority of the debris accumulated in the area between the toe of
- 13 -
the slope and the railway, though a number of boulders and approximately 1 m3 to 2 m3 of finer
material crossed the up-line track 3 m away from the toe of the slope. Some debris was also
retained in the rapture area. The travel angle of the landslide debris, measured from the crest of
the failure to the distal end of the main body of the landslide debris, was about 38° (Figure 6).
This angle is within the typical range for rain-induced soil cut slope failures in Hong Kong
(Wong & Ho, 1996), indicating that the debris mobility of this landslide is comparable to that
commonly observed.
Obstruction of the railway tracks by the debris disrupted the normal KCRC service
between 07:53 hours and 10:39 hours. Maximum north-bound and south-bound service gaps of
92 minutes and 98 minutes were recorded respectively. Disruption to the service was
aggravated by another failure (GEO Incident No. ME97/7/30) on the same morning near Fo Tan
Station to the south of University Station. The landslide is the subject of a separate detailed
study report.
4.
RAINFALL
The nearest GEO automatic raingauge No. N09 is located at the Meteorology
Laboratory at Chinese University, about 400 m northwest of the site (Figure 1). The daily
rainfall recorded by the raingauge between 1 June 1997 and 6 July 1997 together with hourly
rainfall between 29 June 1997 and 2 July 1997 are presented in Figure 7a and 7b respectively.
Isohyets of rainfall recorded between the onset of the storm on 2 July 1997 and 06:10 hours
are shown in Figure 8.
Rain began to fall at about 03:00 hours on 2 July 1997 and intensified significantly
between 05:00 hours and 07:00 hours when 164.5 mm of rainfall was recorded. The 12-hour
and 24-hour rainfall prior to 06:10 hours were 130 mm and 151.5 mm respectively. The
maximum rolling hourly rainfall was 72 mm, recorded between 04:50 hours and 05:50 hours on
2 July 1997. A total of 864 mm of rainfall was recorded by GEO raingauge No. N09 in the
31 days before the failure.
Table 1 presents estimated return periods for the maximum rolling rainfall for selected
durations based on historical rainfall data at the Hong Kong Observatory (Lam & Leung, 1994).
The 31-day rainfall was the most severe with a corresponding return period of about 8 years.
The historical maximum hourly rainfalls recorded in each month at GEO raingauge
No. N09 are shown in Figure 9. The figure shows that the maximum hourly rainfall of 94.5 mm
for July 1997 was recorded between 06:00 hours and 07:00 hours on 2 July 1997, though
86.5 mm of this fell after the landslide had occurred. Prior to the landslide, the maximum
clock-hourly rainfall in July 1997 was 70 mm, recorded between 05:00 hours and 06:00 hours
on 2 July 1997 (Hong Kong Observatory, 1997). This is the third highest on record, with
rainstorm events in April 1989 and October 1996 recording greater maximum hourly rainfalls.
- 14 -
5.
The close correlation between the rainfall event of 2 My 1997 and estimated time of
failure indicates that heavy rainfall probably triggered the landslide.
Three possible sources of water may have contributed to the landslide:
(a) surface storaiwater run-off from Tai Po Road,
(b) leakage from subsurface drainage and water carrying services
in Tai Po Road, and
(c) direct rainfall on and adjacent to the landslide site.
Stormwater run-off from Tai Po Road to the landslide is considered unlikely since the
design of the road drainage is such that surface water on the road falls away from the direction
of the failed slope. Furthermore, there was no evidence of any significant surface water flow on
the slope between the road and the landslide area.
Throughflow beneath the road, perhaps involving leakage from the subsurface drainage
system in Tai Po Road was also discounted as there were no signs of such throughflow such as
piping. It is also unlikely that significant quantities of water could travel underground from the
drains located in the inside kerb of Tai Po Road to the failure site, a distance of some 30 m, in
the period between the on-set of the storm (04:00 hours) and the time at which the failure was
first recognised (06:08 hours). There is no evidence that water-carrying services at the crest of
the slope were leaking prior to the failure.
Direct rainfall on and adjacent to the landslide is, therefore, the most probable source of
water involved in the failure.
The failure was very shallow, and involved loose, apparently undisturbed residual soil,
and underlying completely decomposed granite. A planar surface in weathered rock was
exposed in much of the base of the rupture surface.
Two distinct possible causes of failure related to direct rainfall on and adjacent to the
landslide area are considered:
(a) localised erosion by washout action of surface water, and
(b) infiltration of water into the unprotected slope, leading to
increased bulk unit weight and reduced shear strength along
the failure surface.
Washout action by surface water flow is considered an unlikely mechanism, in view of
the lack of evidence of storm water run-off from Tai Po Road and absence of features that might
concentrate and direct flow over the slope.
- 15 -
Infiltration into the unprotected slope, therefore, is the most probable cause of the
landslide, but the actual mechanism of failure remains unclear. Probably, a number of factors
contributed, notably increased bulk unit weight of the overburden due to saturation, loss of
suction, and development of positive water pressures. It is likely that the planar surface
exposed in the failure scar was a pre-existing discontinuity and as such may have had a
slightly lower shear strength than the surrounding completely decomposed granite.
Furthermore, critical water pressures possibly developed along this surface, thereby reducing
the available shear strength and enabling failure to occur. It is not clear why the slope did not
fail during previous heavy storms in 1989 and 1996. The possibility that local stability of the
slope had deteriorated with time cannot be excluded. The comparatively wet month that
preceded the landslide may have contributed to wetting of the slope, making it susceptible to
failure in response to the short-term rainfall of relatively low severity that occurred.
There are no known links between the previous failure in 1980, on the lower cut slope,
and the recent failure of 1997, other than their proximity. It is possible that they might both
reflect a local adversely oriented joint network.
6.
CONCLUSIONS
It is concluded that the landslide was principally caused by infiltration of water into
the unprotected slope following a period of intense rainfall. The presence of a relict adverse
discontinuity in the weathered rock may have also contributed to the failure due to possible
lower shear strength and development of localised water pressures at the discontinuity.
7.
REFERENCES
Bachy Soletanche Group (1986). KCRC - Geotechnical Investigation of Existing Slopes Stage 2. Ground Investigation Factual Report, vol. 2, Bachy Soletanche Group,
Contractors to KCRC, 1986.
Binnie & Partners (1978). Landslide Studies Phase 1 Re-Appraisal of Cut Slopes and Natural
Slopes and Retaining Walls. Slope No. 7NE-C/C106. Technical Report. Binnie &
Partners (HK) Limited, Consulting Engineers for Government of Hong Kong, 2 p.
Geotechnical Control Office (1986). Shatin: Solid and superficial geology. Hong Kong
Geological Survey. Map Series HGM 20, Sheet 7, 1:20 000 scale. Geotechnical Control
Office, Hong Kong.
Geotechnical Control Office (1987). Geotechnical Area Studies Programme - Central New
Territories. Geotechnical Control Office. Hong Kong, GASP Report no. II, 165 p.
plus 4 maps.
Geotechnical Control Office (1988). Feature Reference No. 7NE-C/C106, University
Railway Station (South End). Stage 1 Studies Summary Report. Geotechnical
Engineering Office, Hong Kong, 1 p.
- 16 -
Hong Kong Observatory (1997). Rainfall Record, Hong Kong Observatory Correspondence to
HAP Reference HKQG21/50. dated 12 September 1997, 2 p.
KCRC (1997a). Landslide Incident at KCRC, Slope near University Station. KCRC
Correspondence to HAP Reference ERD/1BWN/PJT/01 dated 22 August 1997.1 p.
KCRC (1997b). Notes of Meeting For Preventive Measures - Landslide. Notes of KCRC
Meeting on 23 July 1997. 6 p.
KCRC (1997c). KCRC Slope Inspection Data Sheet Slope No. U14.3, May 1997, 9 p.
KCRC (1998). KCRC Incident Report Form (File No. 970098). KCRC Correspondence to
HAP dated 4 February 1998. 3 p.
King, J.P. (1994). Guidelines for Phase 2 SIFT (Version 3.1). Internal Report. Geotechnical
Engineering Office, Hong Kong, 66 p., plus 5 addenda.
Lam, C.C. & Leung, Y.K. (1994). Extreme rainfall statistics and design rainstorm profiles at
selected locations in Hong Kong. Royal Observatory. Hong Kong. Technical Note
no, 86, 89 p.
Ove Arup & Partners (1984). Kowloon-Canton Railway Geotechnical Study - Initial
Assessment and Preliminary Report. Technical Report, Ove Arup and Partners Limited,
Consulting Engineers for KCRC.
Ove Arup & Partners (1987a), Geotechnical Investigation of Existing Slopes Report on Stage
2 Investigation, Executive Report. Technical Report. Ove Arup and Partners Limited,
Ove Arup & Partners (1987b). Geotechnical Investigation of Existing Slopes Report on Stage
2 Investigation, Analysis. Technical Report. Ove Amp and Partners Limited,
Wong, H.N. & Ho, K.K.S. (1996). Travel distance of landslide debris. Proceedings of the
Seventh International Symposium on Landslides, Trondheim, Norway, vol 1
pp 417-422.
- 17 -
LIST OF TABLES
Table
No.
1
Page
No.
Maximum Rolling Rainfall at GEO Raingauge No. N09
for Selected Durations Preceding the 2 July 1997
Landslide and The Corresponding Estimated Return Periods
18
- 18 -
Table 1 - Maximum Rolling
at GEO
No. N09 for Selected
Durations Preceding the 2 July 1997 Landslide and The Corresponding
Estimated Return Periods
Duration
Maximum Rolling
Rainfall (mm)
End of Period
Estimated Return
Period (Years)
5 minutes
12
05:40 hours on 2 July 1997
2
15 minutes
31.5
3
1 hour
72
2
2 hours
90
2
4 hours
117
2
12 hours
122
1
24 hours
144.5
1
2 days
207
2
4 days
234.5
1
7 days
310.5
2
1 5 days
476
2
31 days
925
8
Notes:
(1)
Return periods were derived from the Gumbel equation and data
published in Table 3 of Lam & Leung (1994).
(2)
Maximum rolling rainfall was calculated from 5-minute data for
durations up to one hour and from hourly data for longer rainfall
durations.
- 19 -
LIST OF FIGURES
Figure
No.
Page
No.
1
Site Location Plan
20
2
Plan of the Landslide Site
21
3
Land Status and Site History
22
4
Cross-section A-B through the Southern Part of Cut Slope
No. 7NE-C/C106 Showing the Presence of the Fill Body
23
5
Site Plan and Landslide Details
24
6
Cross-section through the Landslide
25
7
Rainfall Records at GEO Raingauge No. N09
26
8
Isohyets of Rainfall between 03:20 Hours and 06:10 Hours
on 2 July 1997
27
9
Maximum Hourly Rainfall (mm) Recorded in Each Month at GEO
Raingauge No. N09 from 1984 to 1997
28
- 20 -
/ New Territories
GEdxRaingauge No. N09
t
N
Key Plan
Ma Liu Sfaui
Ferry Pier
GEO Ralngauge No. N09
830600 M
Tolo Highway
Landslide Site
cl
Cut Slope No
7NE-C/CI06
/ ;
50
i
Scale
Note :
Base map is extracted from Survey Sheet Nos. 7NE-22A, 7NE-22B, 7NE-22C
& 7NE-22D dated October 1992 (Original scale 1:1 000).
Figure 1 - Site Location Plan
100m
I
- 21 -
Subsurface road
dramage system
100 mm diameter
metal pipe at surface
Exposed landslide scar
Cut Slope No
7NE-C/C106
Debris retained
in landslide scar
Approximate extent
of debris
KCRC north-bound track
i
t
Upslope edge of
^ Dip direction of planar slope
shotcreted area
° and inclination
Orientation of
60 n Dip direction of undulating slope
photograph and Plate No
and inclination
Based on available topographic map and site inspections
30
Figure 2 - Plan of the Landslide Site
Break of slope
- 22 -
o
cs
cd
OQ
GO
a
cd
m
0)
bO
23 -
O
G
z
<u
Ou
ô
CO
>-J
&
U
c+-<
O
>
JC3 ^
t± C
^
QJ
JU
QJ
-2
g^
- S
QQ ^
i ^
< %
«^
O O)
"43
^
o -*-
- 24 -
,\
Fence
Vertical plastic tube in
ground, probably borehole
BH18 of KCRC's 1987 investigation
{Bachy Soletanche Group, 1986)
Exposed planar surface
of completely decomposed
fine- to medium-grained granite
Approximate extent of
debris retained In the
rupture area
Granite residual soil
0.6 m - 1 m thick
Minor I surficial
erosion and gullying
Concrete
retaining structure
Approximate position
of debris
Scale
Legend:
Notes:
3Q,\
Direction of planar slope
Break of slope
and inclination
1 60° Direction of undulating slope
Location of cross-section
and inclination
shown in Figure 6
(1) Based on information recovered from site inspections.
(2) Refer to Figure 6 for more detailed soil descriptions.
Figure 5 - Site Plan and Landslide Details
- 25 -
"O
c
C/3
T3
O
C *£
b
w>
n
ii
2 en<E>
a
<D
J-i
C3
c o
-o
Si
3
3
J2
o
x.
DO
o
W3 OH
1is °1
03
Qj>
> -o
O
$-1
o
• *H
HM
•
VO
W O
cti j-
g
g
^
- "S-s &«
& §£
O
£
CD »-(
fc- {jfl
Q
ZH
3s
I
•
C3
0)
O ^^ .S -5
G
O
ao
CO
O
J-i
O
i
"Q
MD
<L>
*J
+*
O<
o
C!
x u£ ° 5
J>
,—I
^H
fl>
S 2 j£ •5
0
_ 5 w> o c3
S
Q
tH
ffl
CM
G
O
O
~~*
O
*-«
® in *^
"2 5 S »n §
,2
Q> »_ ^5
^-* 0
j;, o»-«
(Tj
"r
Q
CJ
r-<
£)£)
o
"2
^ _D •-'
g x
rS bD Q>
S
^ ^ ;^ 53
*-<
G
»M O PH
3
^
oo
g
O
<u
OT
-rt
w
-*-* «J
^ ^
a| |
o
X~N
/^S
/^~\
X"-\
*-H
<N
CO
Tf
5
- 26 -
|
864 mm of rainfall recorded in the 3 1 days before the time that the landslide was reported
^
488 5 mm of rainfall recorded in the 15 days
before the time that the landslide was reported^
900 -i
800 I .mirfslide occurred at about
700 ^
J=
600-
g
500 -
.s
*«s
Od
400 -
Q
300-
.^
06 08 hours on 2 July 1997
200 -
10001
ii.
3
5
i. lf
7
9
11
13
15
17
, T I.I T ...|
19
21
23
25
27
29
1
June 1997
3
5
July 1997
(a) Daily Rainfall Recorded Between 1 June and 6 July 1997
151 5 mm of rainfall recorded in the 24-hours before the landslide
130 mm of rainfall recorded in the 12-hours before the landslide
L\JV ~
x-x
90 -
Peak hourly rainfall recorded between
80-
06 00 hnnrs and 07 00 hnnr« Note
landslide occurred at about 06 08 hours
^
70-
^ 601
£
o
40-
ffi
302010 00 00
12 00
29 June 1997
000
, i
1200
30 June 1997
000
ill.....
1200
1 July 1997
ii i 1
I
1200
000
2 July 1997
(b) Hourly Rainfall Recorded Between 29 June and 2 July 1997
Figure 7 - Rainfall Records at GEO Raingauge No. N09
000
27 -
o
en
0
O
o
C/3
^3
O
ffi
O
^
o
a
<D
<D
4-»
^
<D
C4H
O
oo
o
CO
i—i
i
oo
<D
00
- 28 -
=
——.sj
1
*o
l>
ON
ON
ON
ON
ESasasgjEjj
2
TT
00
ON
£
"
S
c§
O
t3 T3
ON Q**
, g g
ON
0
»„
BMit^MiaaaaBai,
o;-
ON
ON
0*^0
•& j3 g
«D
ON
ON
s
^ -a -8
*O"
So
O
*S
.S
|"g^
^ « |
-_^__ ___^J
^ga
S
ON
"H
6
Z
••••j!
en
^1e
/L,
bD
cd
1
o
«
IMIIBMIIIIIIIil
g-|§
2
-3
S
o
i^H
s § *H £
^s §•§
„.—
1 5 xx:
2 g
2
.^^^.^^^.^j
O
ON
ON
.3
-d<D
nM^M^S
oo
ON
T3
ON
O
00
--— -^^--^ -g-gi^.—
1
J-H
o
rts
/—s
OO
OO
ON
ESS
r-
-g-2
00
1
ON
p^>
ON
OO
£ Io
as-
,-»
mamsm
1
MH
laximum
OO
&
OS
<D
«
1
• rH
fe
i1
§
8
§
(uiui) [{BjuiFg XpnoH umuiixBp\[
S
>
- 29 -
LIST OF PLATES
Plates
No.
Page
No.
1
General View of the Landslide Site (Photograph Taken on 4 July 1997)
30
2
General View of the Northern Flank of the Landslide Area
(Photograph Taken on 4 July 1997)
31
3
General View of the Southern Flank of the Landslide Area
32
4
Copy of Photograph Taken by KCRC Staff Soon After Failure on
2 July 1997 (Photograph Taken on 2 July 1997)
33
- 30 -
Plate 1 - General View of the Landslide Site
31 -
t>
3
~
I
~r
S
-.
p.
:
:
D
£
H
:
i
4n
O
I
i
CO
-I
- 33 -
Plate 4 - Copy of Photograph Taken by KCRC Staff Soon After Failure
on 2 July 1997 (Photograph Taken on 2 July 1997)
- 34 -
SECTION 2 :
DETAILED STUDY OF THE
LANDSLIDE NEAR
KOWLOON-CANTON RAILWAY
CORPORATION FO TAN
STATION ON 2 JULY 1997
This report was originally produced in March 1998
as GEO Landslide Study Report No. LSR 4/98
- 35 -
FOREWORD
This report presents the findings of a detailed study of a
landslide (GEO Incident No. ME97/7/30) which occurred on
2 July 1997, about 250 m southwest of the Kowloon-Canton Railway
Corporation (KCRC) Fo Tan Station. Debris from the landslide
obstructed the north-bound KCRC rail track. The first north-bound
train of the day was partly derailed by the landslide debris, causing
significant disruption to rail services. No fatalities or injuries were
reported.
facts about the landslide, present relevant background information
and establish the probable causes of the landslide. The scope of the
study was generally limited to site reconnaissance, desk study and
analysis. Recommendations for follow-up actions are reported
separately.
report was written by Mr M Riley and reviewed by Dr R Moore and
Mr H Siddle. The assistance of the GEO in the preparation of the
report is gratefully acknowledged.
G. Daughton
Project Director/Halcrow Asia Partnership Ltd
- 36 -
CONTENTS
Page
No.
Title Page
34
FOREWORD
35
CONTENTS
36
1.
INTRODUCTION
37
2.
THE SITE
37
2.1
Site Description
37
2.2
Site History
38
2.3
Previous Studies
39
3.
THE LANDSLIDE
39
3.1
Time of Failure
39
3.2
40
3.3
41
4.
RAINFALL
42
5.
42
6.
CONCLUSIONS
43
7.
REFERENCES
44
LIST OF TABLES
45
LIST OF FIGURES
48
LIST OF PLATES
59
- 37 -
1.
INTRODUCTION
On the morning of 2 July 1997, a landslide (GEO Incident No. ME97/7/30) occurred
on a slope about 250 m southwest of the Kowloon-Canton Railway Corporation (KCRC) Fo
Tan Station, Sha Tin (Figure 1). Debris from the landslide obstructed the north-bound KCRC
rail track. The first north-bound train of the day was partly derailed by the landslide debris,
causing significant disruption to rail services throughout the morning. No fatalities or injuries
were reported.
Following the landslide, Halcrow Asia Partnership Ltd (the 1997 Landslip
Investigation Consultants) carried out a detailed study of the failure for the Geotechnical
Engineering Office (GEO), Civil Engineering Department (CED) under Agreement
No. CE 68/96. This is one of a series of reports prepared during the consultancy by Halcrow
The key objectives of the study were to document the facts about the landslide, present
relevant background Information and establish the probable causes of the failure. The scope
of the study was limited to site reconnaissance, desk study and analysis. Recommendations
for follow-up actions are reported separately.
key tasks:
site,
(c) discussions with KCRC staff and other parties who
inspected the incident,
(d) detailed site observations and measurements at the
landslide, and
(e) diagnosis of the probable causes of failure.
2.
THE SITE
2.1
Site Description
A plan of the landslide location is shown in Figure 2. The slope that failed was a 26 ni
high, densely vegetated, partly modified natural hillside (Plate 1). At the crest of the slope
there was a 3 m wide concrete footpath, which provides access to Sui Wo Court, uphill and
northwest of the landslide. The footpath was positioned on the southern edge of a cut
platform, which extends a further 20 m to the north. A network of surface drains has been
installed on the platform (Figure 2).
At the location of the landslide the slope faces southeast and was inclined generally at
45°. The inclination of the lowest 8 m of the slope was about 55° to the horizontal and near
- 38 -
the crest, the slope angle reduces to 25°. The area between the toe of the slope and the KCRC
track was approximately level.
An abandoned squatter platform exists mid-way up the slope which was accessed
down a concrete stairway from the main footpath at the crest of the slope (Plate 3). A squatter
village is located southwest of the landslide.
The landslide affected land within the KCRC boundary, but also extended into
unallocated Government land and Sha Tin Town Lot (STTL) No. 43sA (Figure 2). STTL
No. 43sA is currently under the responsibility of the Housing Department.
2.2
Site History
The history of development of the site was determined from a review of aerial
photographs, available documentary records and information contained in GEO's Landslide
Incident Report database. Key findings are summarised in Table 1 and shown on Figure 3.
Shallow excavations were evident at the toe of adjacent slopes in 1963 (Table 1). It is
possible that the lower, steep section of slope at the landslide location had been modified by
similar excavations.
A linear body of fill was observed on the 1980 aerial photographs immediately
adjacent to the footpath at the crest of the slope, which was probably placed during the
construction of the footpath and platform at about that time.
The GEO's Landslide Incident Report database indicates that three minor failures
occurred nearby in July 1987 (Incident Nos. ME87/7/18, 18A and 18B, Figure 3).
Information available in the database indicates that the volume of each of the failures was
about 3 m3 and that a nearby squatter hut was affected. The incidents were reportedly caused
by infiltration. A failure (Incident No. ME93/5/2) also occurred near a footpath in the same
village in May 1993 (Figure 3). The volume of failure was reported as 6 m3 and the failure
was attributed to infiltration and an oversteep slope profile.
A squatter hut (No. XRCSW/13), previously located on the platform affected by the
landslide on 2 July 1997, was permanently evacuated following a failure on 11 June 1993
(Incident No. ME93/6/7, Figure 3). According to the GEO Incident Report, the failure was
about 2 m3 in volume and occurred on a minor cut slope formed behind the squatter hut. The
failure involved weathered rock and was attributed to infiltration.
Two small landslides (GEO Incident Nos. ME97/7/98 and ME97/7/110), located
immediately southwest of the July 1987 failures (Figure 3), were reported to the GEO on
16 July 1997 and 24 July 1997 respectively. Shotcrete was applied to the slopes soon after
the failures and this prevented detailed inspection by HAP. According to the GEO Incident
Reports, the failures each involved about 2 m3 of residual soil and were caused by infiltration.
The KCRC rail tracks were not affected.
- 39 -
2.3
Previous Studies
The slope that failed was a substantially unmodified natural slope in the late 1970's,
and was not registered in the 1977/78 Catalogue of Slopes by the consultants engaged by the
Government to prepare the Catalogue.
In 1984, KCRC began a comprehensive geotechnical investigation of existing manmade slopes adjacent to their rail tracks (Ove Amp & Partners (OAP), 1987). The project
included preliminary (Stage 1) cataloguing and assessment of all "cutting and embankment
slopes along the railway between Hung Horn and Lo Wu" (OAP, 1987). The Stage 1
assessment involved field inspection and stability analysis based on assumed soil parameters
and was intended to identify potentially unstable features for further (Stage 2) study. The
findings of the assessment were presented in a report prepared in 1984, but this could not be
found in KCRC's records. It has not been possible, therefore, to establish whether the slope
affected by Incident No. ME97/7/30 was specifically included in the Stage 1 assessment. The
Stage 2 Studies were based on a programme of site specific ground investigation and
laboratory testing and the results were presented in a report prepared in April 1987 (OAP,
1987). An assessment of the affected slope was not included in the Stage 2 investigation
report.
In 1992, the GEO initiated a consultancy agreement entitled "Systematic Inspection of
Features in the Territory" (SIFT) which aims, inter alia, to systematically search for slopes not
presently included in the 1977/78 Catalogue of Slopes for inclusion in the New Catalogue of
Slopes. The SIFT records for the area indicate that the part of the slope affected by Incident
No. ME97/7/30 was not identified for registration in the New Catalogue.
KCRC carry out annual inspections of slopes adjacent to their rail tracks (KCRC,
1997a). The part of the slope affected by Incident No. ME97/7/30 forms the southern part of
Slope No. Ul 1.6(6.2) in KCRC's system. According to KCRC's records, Slope
No. Ul 1.6(6.2) was last inspected on 23 June 1997 shortly before the landslide. A number of
observations and recommendations were made, but most notably "slumping and erosion" at
locations to the north and south of the landslide site were reported (Figure 3). The date and
time of the slumping and erosion, however, was not known. Following the inspection it was
recommended to "invite geotechnical specialists to assess the stability of the slope" (KCRC,
1997b). After the 2 July 1997 failure, KCRC instigated a geotechnical review of existing
slopes within their boundaries and adjacent to the railway tracks. Slope No. Ul 1.6 (6.2) was
included for assessment in the review.
3.
THE LANDSLIDE
3.1
Time of Failure
A report prepared by KCRC following the landslide incident (KCRC, 1997c) noted
that the derailment of the train was confirmed at 05:54 hours on 2 July 1997. The last northbound train passed unobstructed through Fo Tan Station at about 00:40 hours on 2 July 1997.
The time of failure, therefore, may reliably be constrained to between 00:40 hours and
05:54 hours on 2 July 1997. A landslide warning was raised at 06:25 hours on 2 July 1997.
- 40 -
3.2
Details and observations of the landslide made during site inspections are given in
Figures 4 and 5. Cross-sections of the landslide are presented in Figures 6 and 7 and
photographs are shown in Plates 1 to 6.
The landslide originated from the upper part of the slope. It produced a failure scar
typically about 15 m long and 8 m wide, with an average depth of about 1.5 m. The volume
of failure was estimated to be about 100 m3 (Plate 1).
The shape of the main scarp of the landslide suggests that there might have been two
phases of movement. The initial failure, a translational debris slide, occurred immediately
north of the squatter platform. It was the larger of the two phases, with an estimated volume
of about 80 m3. The debris travelled down the slope and crossed the north-bound tracks, a
distance of about 15m from the toe of the slope (Plate 2). The travel angle of the debris from
this failure phase, measured from the crest of the failure to the distal end of the main body of
the debris, was about 30° (Figure 6). This angle is within the typical range for rain-induced
landslides in Hong Kong (Wong & Ho, 1996), indicating that the debris mobility from this
failure is comparable to that commonly observed. The failed mass observed at the toe of the
slope was almost totally disintegrated and comprised pale-orange, coarse-grained silty sand
with occasional cobbles and boulders (Plate 2). The material was very wet and was deposited
in a layer between 0.1 m and 0.3 m thick. The squatter platform part way up the slope was
partly demolished by the initial failure (Plate 3). The failure also undermined the ground
further up the slope, below the footpath to Sui Wo Court, which was involved in the
subsequent failure phase. A plastic pipe, approximately 100 mm in diameter was observed
immediately behind the main scarp of the initial failure (Figure 5, Plate 4). No water was
seen issuing from the pipe during inspection by HAP on 4 July 1997.
The second phase failure was also a translational debris slide, involving about 20 m3
of material near the crest of the slope below the footpath. The failed mass was less mobile,
with most of the debris remaining as a disturbed raft within the rupture area (Figure 5). The
travel angle of the debris from this failure phase was 38° (Figure 7), again within the typical
range for rain-induced landslides in Hong Kong (Wong & Ho, 1996).
KCRC staff, when inspecting the landslide on the day of the failure, observed surface
water flowing along the platform at the crest of the slope (Figure 4). Part of the surface flow
was channelled into the landslide by the footpath and staircase leading to the squatter
platform. The washout action of the surface flow eroded a gully below the rupture zone of the
initial failure (Plate 4) and significantly reworked debris at the toe of the slope.
Following the landslide, loose debris was removed by KCRC and shotcrete applied to
the surface of the slope at the recommendation of the GEO. Weepholes were installed at
regular intervals to enable drainage of water from the slope. The GEO also advised Buildings
Department that a Dangerous Hillside Order (DHO) be served with respect to the failed slope
on private land, and that the Design Division of GEO investigate and, if necessary, upgrade
adjacent areas on Government land.
- 41 -
3.3
Geological observations made on site are summarised in Figures 5 to 7.
The failure occurred principally in weathered, coarse-grained granite. The weathered
rock, however, was overlain at the eastern part of the landslide by about 1 m of brown sandy
soil material, which may be fill, and a thin veneer of topsoil. A thin layer of coarse gravel,
cobble and boulder size fragments of highly decomposed granite, also probably fill, mantled
weathered rock in much of the southwest side of the failed area. These materials may
correspond to the fill body which is believed to have been formed during construction of the
platform at the crest of the slope in 1980, or may be associated with construction of the
squatter platform and access stairway.
The landslide exposed slightly to moderately decomposed granite in the lower part of
the slope, and highly and completely decomposed materials higher up. Heavily stained
surfaces dipping at between 45° and 55° were exposed locally in the base of the landslide scar,
which were interpreted as sheeting joint surfaces (Figure 5).
A particularly pronounced surface, inclined at 45°, was exposed in the area associated
with the initial failure. Two voids or erosion pipes, typically between 50 mm and 70 mm in
diameter, extended into the slope at the interface between a similar exposed surface and
overlying material near the central part of the rupture zone (Plate 6). Soil around the pipes
and on the surface was heavily stained.
A prominent pervasive wavy joint surface observed in the face of the lower part of the
slope below the landslide scar, was inclined out of the slope at between 26° and 46°. The
wavelength of the undulation was typically between 4 m and 6 m, and the amplitude about
1 m. The surface was interpreted as a sheeting joint. Similar surfaces occur throughout the
full height of the slope at spacings between about 0.1 m and 4 m. The surfaces exposed in the
main landslide scar also appear to belong to the sheeting joint system. The joint surfaces
lower in the slope are locally heavily stained, containing up to 100 mm of closely fractured,
completely decomposed granite. Seepages from joints near the toe of the slope were observed
after the failure (Figure 5). Flows of between about 2 1/min. and 3 1/min. were estimated
during an inspection on 26 September 1997.
Two sub-vertical, planar joint sets dipping typically towards 070° and 320°, at
spacings of between 0.5 m and 2 m were evident in the lower part of the slope and exposed
locally in the main scarp. These surfaces interact in places with the sheeting joint set, and
may have provided release surfaces in the rupture area.
Sheet 7 of the Hong Kong Geological Survey 1: 20 000 Scale Map (Geotechnical
Control Office (GCO), 1986), and the 1987 Geotechnical Area Studies Programme (GASP)
Engineering Geology Map for the Central New Territories (GCO, 1987), indicate that the site
is underlain by coarse-grained granite. The GASP map also indicates a linear fill body
orientated close to the alignment of the footpath at the crest of the slope. These findings are
in general agreement with observations made on site.
- 42 -
4.
RAINFALL
Analysis of rainfall records was carried out to assess the influence of rainfall on the
landslide. The analysis was based on rainfall readings taken at 5-minute intervals by a
network of automatic raingauges located throughout Hong Kong and transmitted to HAP by
the Hong Kong Observatory.
The nearest GEO automatic raingauge No. N02 is located at Shun Wo House, Wo Che
Estate, Sha Tin, about 650 m southwest of the landslide. Figure 8 shows the location of
raingauge and presents isohyets of rainfall recorded between 18:00 hours on 1 July 1997 and
06:00 hours on 2 July 1997. Daily rainfall recorded at the raingauge from 2 June to
9 July 1997 and hourly rainfall between 30 June and 3 July 1997 are shown in Figures 9a and
9b respectively.
Rainfall on 2 July 1997 started at about 01:00 hours and intensified significantly
between 05:00 hours and 06:00 hours. The maximum rolling hourly rainfall (92.5 mm) was
recorded between 04:55 hours and 05:55 hours. The 12-hour and 24-hour rainfall totals
recorded prior to 05:55 hours on 2 July 1997 were 68 mm and 138.5 mm respectively.
Table 2 presents estimated return periods of maximum rolling rainfall for selected
durations based on historical rainfall at the Hong Kong Observatory (Lam & Leung, 1994).
The maximum rolling 31-day rainfall was the most severe with a corresponding return period
of about 8 years. The maximum rolling 1-hour rainfall intensity was also notable, having an
estimated return period of about 6 years.
Historical maximum hourly rainfall totals recorded in each month at raingauge
No. N02 since installation are presented in Figure 10. Figure 10 indicates that the maximum
hourly rainfall for July 1997 (99 mm recorded between 05:00 hours and 06:00 hours on
2 July 1997), was associated with the landslide and may be among the highest recorded.
5.
The landslide comprised principally two phases of sliding, though significant
secondary erosion by surface water flow also occurred. The sliding surface of the initial
failure phase was controlled by an adversely oriented sheeting joint surface. Sub-vertical
joint systems were also observed in the main scarp of the initial failure. The sub-vertical and
sheeting joint surfaces were probably significant in causing the failure, by enabling
detachment of the failure mass and preferential flow and accumulation of water.
The close correlation between the rainfall event of 2 July 1997 and the likely time of
failure indicates that heavy rainfall most probably triggered the landslide. Four possible
sources of water at the landslide site were considered, namely:
(a) leakage of water from the pipe observed in the main scarp of
the failure,
(b) direct infiltration of rainfall into the slope,
- 43 -
(c) saturation of near surface materials by surface water
originating from the platform and stairway, and
(d) subsurface seepage to the landslide area through the sheeting
joint system.
There was no evidence to Indicate that the plastic pipe observed near the main scarp of
the initial failure was leaking before the failure and It seems unlikely that this was a
significant source of water. Infiltration of rainfall and surface run-off from the footpath and
stairway would both have enabled water to enter the slope, assisted by the absence of surface
protection and the localised presence of fill and sub-vertical joints. However, rainfall before
the failure, though heavy, was not exceptional with calculated maximum return periods
between about 6 and 8 years. This may indicate that infiltration of surface run-off was more
influential in causing the failure than direct infiltration of rainfall.
The heavily stained surfaces and localised erosion pipes observed in the main scarp
also indicate that periodic subsurface water flow occurs along the sheeting joints. It is
possible that subsurface water, transmitted through the sheeting joints, also reached the
landslide site from the upslope area.
Water entering the slope by the above mechanisms probably caused an increase of
water pressure in the soil mass and enabled build-up of local water pressure along pre-existing
discontinuities in the weathered granite. The increase in water pressure would have resulted
in reduction of the available shear strength, principally along the main sheeting joint surface,
leading to the initial failure near the squatter platform. Infiltration of rainfall into the nearsurface materials would also have increased their bulk density, and contributed further to the
de-stabilising effect on the soil mass. Undermining of the upper parts of the slope by the
initial failure was the principal cause of the second failure phase.
The hillside has been modified by the formation of a platform and placement of fill at
its crest, construction of a squatter hut and access stairway and, possibly, by shallow
excavations at its toe. These modifications probably only marginally reduced the overall
factor of safety of the hillside. However, the footpath associated with the platform at the
crest, together with the abandoned squatter hut and stairway, may have served to concentrate
and discharge surface water flows into the landslide area. It is possible that the fill at the crest
of the slope also contributed by locally promoting infiltration.
The significance of the slumping and erosion of the slope noted before the landslide
during an inspection by KCRC on 23 June 1997 Is unclear, but observation of these features
may indicate that other parts of the slope are also subject to periodic surface water flows,
presumably during times of heavy rainfall.
6.
CONCLUSIONS
It is concluded that the landslide south of Fo Tan KCRC Station on 2 July 1997 was
triggered by heavy rainfall. The probable cause of failure was build-up of transient water
pressure within the soil mass and along an adversely oriented sheeting joint system. The
increase in water pressure would have reduced the available shear strength and enabled failure
to occur. Surface water flowing onto the slope from above the crest area, together with heavy
- 44 -
rainfall on the slope surface, were probably the most significant sources of water. Subsurface
flow along the sheeting joint in response to heavy rainfall, however, may have also
contributed to the generation of transient water pressure on the joint surface. Saturation of the
near-surface materials would also have increased their bulk density and contributed to the destabilising effect on the soil mass.
7.
REFERENCES
Geotechnical Control Office (1986). Sha Tin: Solid and superficial geology. Hong Kong
Geological Survey, Map Series HGM 20, Sheet 7, 1:20 000 scale. Geotechnical
Control Office, Hong Kong.
Territories. Geotechnical Control Office, Hong Kong, GASP Report no. II, 165 p.,
plus 4 maps.
KCRC (1997a). Notes of Meeting for Preventive Measures - Landslide. KCRC Minutes of
Meeting on 23 July 1997. 4 p.
KCRC(1997b). KCRC Slope Inspection Data Sheet: Slope No. U11.6 (6.2). KCRC
Maintenance Inspection Record, dated 23 May 1997, 13 p.
KCRC (1997c). Derailment of H001 (T22) at south of Fo Tan on 2/7/1997. KCRC Internal
Memorandum, dated 2 July 1997.
selected locations in Hong Kong. Royal Observatory. Hong Kong. Technical Note
no. 86, 89 p.
Ove Arup & Partners (1987), Investigation of Existing Slopes. Report on Stage 2
Investigation - Executive Report. Technical Report. Ove Arup and Partners Limited,
Consulting Engineers for Kowloon-Canton Railway Corporation.
Wong, H.N. & Ho, K.K.S. (1996). Travel distance of landslide debris. Proceedings of the
Seventh International Symposium on Landslides. Trondheim, Norway, vol.1,
pp 417-422.
- 45 -
LIST OF TABLES
Table
No.
Page
No.
Key Site Developments Identified from Aerial
Photograph Interpretation and Documentary Review
46
Maximum Rolling Rainfall at GEO Ralngauge No. N02
for Selected Durations Preceding the 2 July 1997
Landslide and The Corresponding Estimated Return Periods
47
- 46 -
Table 1 - Key Site Developments Identified from Aerial Photograph Interpretation and
Documentary Review
Year
Key Development
1949
The hillside appears undisturbed, though areas of shallow excavation are
evident on the lower parts of the slopes.
1963
Almost all of the lower slopes have been excavated, possibly to provide
borrow for reclamation works nearby. Three squatter huts have been built
near the current crest of the slope behind the landslide location.
1977
Squatter huts at crest demolished.
1980
Squatter platform formed mid-way up slope; formation of cut platform at the
crest, associated with minor filling on upper part of slope.
Construction of platform, footpath and drainage system at crest of slope.
1985
Surficial filling below squatter platform.
1987
Three minor failures (GCO Incident Nos. ME87/7/18, 18A and 18B) occurred
near to the village southwest of the landslide site.
1993
Squatter hut (No. XRCSW/13) and platform affected by landslide on
2 July 1997 abandoned following failure of minor cut slope behind building.
1997
Minor failures (GEO Incident Nos. ME97/7/98 and ME97/7/110) reported to
the GEO on 16 July 1997 and 24 July 1997.
- 47 -
Table 2 - Maximum Rolling Rainfall at GEO Ralngauge No. N02 for Selected
Durations Preceding the 2 July 1997 Landslide and The Corresponding
Estimated Return Periods
Duration
Maximum Rolling
End of Period
Rainfall (mm)
Estimated
Return Period
(Years)
5 minutes
11
1
15 minutes
31
3
1 hour
92.5
6
2 hours
116.5
4
4 hours
144.5
3
12 hours
150
2
24 hours
179
2
2 days
227
2
4 days
233
1
7 days
278.5
1
15 days
406.5
2
31 days
909.5
8
Notes:
(1)
Return periods were derived from the Gumbel equation and data
published in Table 3 of Lam & Leung (1994).
(2)
Maximum rolling rainfall was calculated from 5-minute data for
durations up to one hour and from hourly data for longer rainfall
durations.
- 48 -
LIST OF FIGURES
Figure
No.
Page
No.
1
Site Location Plan
49
2
Plan of the Landslide Site
50
3
Site History
51
4
General Site Details and Landslide Observations
52
5
Detailed Landslide Observations
53
6
Cross-section A-C through the Landslide
54
7
Cross-section B-C through the Landslide
55
8
Isohyets of Rainfall between 18:00 Hours on 1 July 1997
and 06:00 Hours on 2 July 1997
56
9
57
Maximum Hourly Rainfall (mm) in Each Month at GEO Raingauge
No. N02 from January 1984 to September 1997
58
10
- 49 -
t
N
Fo Tan KCRC
Station
Key Plan
&284DON; /
82 82 CON ^ s
0
t
10 20 30 m
I
I
Scale
Note :
Base map is extracted from Survey Sheet Nos. 7SE-6A, 7SE-6B, 7SE-6C
and 7SE-6D dated October 1992 (Original scale 1:1 000),
5
- 50 -
CO
0)
00
*O
CD
too
- 51 -
- 52
o
4«1
cti
>
1-4
CD
00
JD
O
00
s
15
Q
GO
13
03
CJ
O
Tf
- 53 -
Tension cracks in destabilised
soil mass above main scarp
Plastic pipe, approx.
100 mm diameter
Abandoned squatter
platform partly
demolished by landslide
Surface water flowlme observed
by KCRC staff soon after failure
Superficial layer (<! m) of
brown sandy soil (Fill?)
Soil below pipeholes
heavily stained and damp
Coarse gravel, cobbles
and boulders of moderately
to highly decomposed
granite in slightly clayey
coarse-grained sand
matrix (probably fill)
Densely vegetated
hillside
Exposed heavily
stained joint surface
Raft of disturbed, vegetated
landslide debris
Exposed, heavily
stained joint surface
Narrow gully formed
by surface water flow
Seepage from sheeting joint
surface estimated at 2-3 I/mm
on 26/9/97. Joint contains up
to 100 mm of extremely
closely fractured completely
decomposed coarse-grained
granite, heavily stained
Landslide debris typically
O.I m to 0,3 m deep
KCRC tracks
Apparent line
Approx. geological
Legend :
Concave
^ of seepage
boundary
of
slope
Landslide
I '- "I Highly decomposed
Fill (?)
main scarp
. Convex change
'
granite
Approx.
distribution
of
slope
Joint, inclined
of landslide debris p'-^jl Completely decomposed
with dip in
Break of slope
24 degrees (long
Approx. distribution * *•** * granite
45
Variable
slope,
of washed/sorted
axis is strike
I
j Slightly to moderately
typical angle
landslide debris
direction)
I
' decomposed granite
Notes :
(1) Base map is extracted from Survey Sheet Nos. 7SE-6C
& 7SE-6D dated October 1992 (Original scale 1:1 000).
(2) Refer to Figure 6 and Figure 7 for cross-sections A-C and B-C.
C
I'
Figure 5 - Detailed Landslide Observations
- 54 -
£
W»—j
U
O o-J
«n
(D
bfi
3
O
v-,
o
"m
O
CD
CO
U
I .-a
sx
g» 1
*««
Q* O
»-. "O
a d
M *3
.2
,<L>
i+H
O
3^
QQ
o> <i> vo
I
o
en
I
o
I
o
CN
- 55 -
toJO
&
O
«-<
X3
-*—»
CJ
PQ
c
o
• 1—I
4~*
o
c
o
£
$
£
o
o
•*-•
4)
co
o
rf
cJ> *: «>
on
O
c;
o>
*j»
c*
s a w
x- w d
.2
o> "«
8 JJ
oo
en
O
lH
u
56 -
«N
a
o
en
*-<
3
o
ffi
o
o
^D
O
cd
s>
ON
a
o
C/3
5-<
^
o
ffi
o
<o
CX)
CD
a>
-f_»
^
CD
o
C/D
4_»
CD
o
c/a
HH
I
oo
CD
*H
3
bfl
- 57 -
899 mm of rainfall recorded in the 31 days before the landslide
396 mm of rainfall recorded in
^_ the 15 days before the landslide
700
Date of the Incident
600 -
500 -
IT
J3,
S'
400 -
300-
Q
200 -
100 -
I
3
5
7
9
11
ia
13
15
17
19
21
23
25
27
29
1
June 1997
IM
3
5
7
July 1997
(a) Daily Rainfall Recorded Between 1 June and 7 July 1997
179 mm of rainfall recorded in the 24-hours prior to 06 00 hours
pgH
pp,
150 mm of rainfall recorded in the 12-hours prior to 06 00 hours
n
i^ft
i. Z.\J
S
Incident Confirmed by
KCRC at 05 54 hours
100 -
1
1
S3
o
ffi
8
°"
60
"
••
40 -
20 -
(Lull
li 1 1
.H
A -
000
L_
«JL_^-
Jl
,
a^L^
800 1600 000 800 1600 000
29 June 1997
30 June 1997
11
JiJ .
_ II
800 1600 000
1 July 1997
(b) Hourly Rainfall Recorded Between 29 June and 2 July 1997
ii
800 1600 000
2 July 1997
58 -
o
ON
ON
O>
oo
ON
in
oo
ON
(utui)
yCpnoH uinuitxBp\[
- 59 -
LIST OF PLATES
Plate
No.
Page
No.
1
General View of the Lower Part of the Landslide (Photograph
Taken on 4 July 1997)
60
2
General View of the Landslide Debris (Photograph Taken on
2 July 1997)
61
3
View from above the Main Scarp (Photograph Taken on
4 July 1997)
62
4
View of the Main Scarp of the Initial Failure (Photograph
63
5
Downslope View (Photograph Taken on 4 July 1997)
64
6
Erosion Pipe Exposed in the Central Part of the Main Scarp
(Photograph Taken on 25 September 1997)
65
- 60 -
Plate 1 - General View of the Lower Part of the Landslide
- 61 -
Plate 2 - General View of the Landslide Debris
- 62-
i
s a
m
- 63 -
- 64 -
'-•
:
-.-
5
E
._
2
3
-
- 65 -
„ ;
Plate 6 - Erosion Pipe Exposed in the Central Part of the Main Scarp
(Photograph Taken on 25 September 1997)
- 66 -
SECTION 3 :
FLOODING INCIDENT
AT MEI CHUNG COURT,
SHATIN ON 2 JULY 1997
This report was originally produced in May 1998
- 67 -
FOREWORD
flooding incident (GEO incident No. MW97/7/7) which occurred on
2 July 1997 at Mei Chung Court, Shatin. During the rainstorm of
2 July 1997, blockage of a natural stream course at a footbridge
caused stream flow to over-spill onto a nearby footpath and into the
Mei Chung Court housing estate. This resulted in severe erosion of
the slopes on the boundary of the estate and within a play area, and the
inundation of the lower level of a multi-storey car park with
floodwater and debris. No fatalities or injuries were reported.
facts about the incident, present relevant background information and
establish the probable causes of the incident. The scope of the study
during the consultancy by Halcrow Asia Partnership Ltd (HAP).
The report was written by Dr P Jennings and reviewed by
Dr R Moore and Mr H Siddle. The assistance of the GEO in the
preparation of the report is gratefully acknowledged.
G. Daughton
- 68 -
CONTENTS
Page
No.
Title Page
66
FOREWORD
67
CONTENTS
68
1.
INTRODUCTION
70
2.
THE SITE
71
2.1
Site Description
71
2.2
Site History
71
2.3
Land Status
73
3.
THE INCIDENT
73
3.1
Description of the Incident
73
3.2
Consequential Damage
74
3.3
Immediate Follow-up Action
75
4.
RAINFALL
75
5.
SOURCES OF FLOOD WATER
76
6.
CONDITIONS AND DAMAGE AT THE ESTATE AND
SURROUNDING AREA
76
6.1
General
76
6.2
Natural Stream Course
76
6.3
Area between the Estate and Footbridge
77
6.4
The Estate
77
7.
PROBABLE CAUSES OF THE FLOODING INCIDENT
78
8.
CONCLUSIONS
78
9.
REFERENCES
79
- 69 -
Page
No.
LIST OF TABLES
80
LIST OF FIGURES
82
LIST OF PLATES
89
APPENDIX A : SUMMARY OF DOCUMENTS REVIEWED
96
- 70 -
1.
INTRODUCTION
This report presents the findings of a detailed study of a flooding incident (GEO
incident No. MW97/7/7) which occurred on 2 July 1997 at Mei Chung Court, Shatin
(Figure 1). During the rainstorm of 2 July 1997, blockage of a natural stream course at a
footbridge caused stream flow to over-spill onto a nearby footpath and into Mei Chung Court
housing estate. This resulted in erosion of the slopes on the boundary of the estate and within
a play area, and the inundation of the lower level of a multi-storey car park with floodwater
and debris (Plate 1). No fatalities or injuries were reported.
Following the flooding incident, Halcrow Asia Partnership Ltd (the 1997 Landslip
Investigation Consultants) carried out a detailed study of the incident for the Geotechnical
No. CE 68/96. This is one of a series of reports produced during the consultancy by Halcrow
The key objectives of the study were to document the facts about the incident, present
relevant background information and establish the probable causes of the incident. The
incident was treated as a special case detailed study, as it primarily involved flooding which
resulted in severe erosion of formed slopes. The scope of the study was restricted to the
geotechnical aspects of the incident. Recommendations for follow-up actions are reported
separately.
key tasks:
site,
(c) interviews with witnesses to the flooding,
(d) detailed observations and measurements at Mei Chung
Court, the surrounding area and the natural stream course,
(e) theoretical analysis of water flow in the natural stream
course,
(f) review of the role and performance of the formed slopes
and associated drainage system at Mei Chung Court during
the flooding incident, and
(g) diagnosis of the causes of the flooding incident.
- 71 -
2.
THE SITE
2.1
Site Description
The location of Mei Chung Court housing estate and surrounding area affected by the
flooding on 2 July 1997 is shown in Figure 2. The flooding affected a paved footpath,
providing access from Mei Tin Road to Fuk Lok Tsuen, and the northern part of the estate
which comprised formed slopes, a play area, the lower level of a multi-storey car park and a
residential block.
The affected slopes were situated on the estate boundary (Figure 2). Slope
No. 7SW-D/C790 was around 10 m high at a gradient of about 35°, and covered in vegetation
with a drainage network of surface channels. On the northern perimeter of the housing estate,
two cut slopes (Nos. 7SW-D/C791 and 7SW-D/CR801) were affected by the floodwater
(Figure 2), and these slopes were around 5 m high at a gradient of about 20°, with a vegetation
cover and a drainage network of surface channels.
Floodwater was diverted from a natural stream, about 80 m north of the estate, where a
branch of the Fuk Lok Tsuen footpath crosses the stream via a footbridge (Figure 2). The
footpath and the deck of the footbridge are constructed in concrete, about 2.5 m wide, and are
used for pedestrian and vehicular access.
The natural stream, the source of the floodwater, passes to the north and east of the
estate before entering the Shing Mun River (Figure 2). The stream is located within a valley
area of some 97 ha, of generally naturally vegetated steep upper hillside, lower gentler slopes
and valley floor containing minor buildings and plantations (Figure 1).
The extent of the drainage catchment feeding water into the stream at the footbridge is
shown in Figure 1. This comprises the lower portion of the valley only, with an area of some
25 ha. A Water Supplies Department (WSD) weir situated across the stream intercepts water
draining from the upper valley. The weir is located some 500 m upstream from the estate and
the associated intake shaft abstracts stream water, via a water tunnel, to Lower Shing Mun
Reservoir for storage or to Shatin Treatment Works for treatment.
2.2
Site History
A review of the site history was undertaken using aerial photographic interpretation
(API) and documentary records. Summary results of the API are presented in Table 1.
The earliest available aerial photographs, from 1949, show the estate site as cultivated
land with several scattered small buildings. The area immediately north of the estate site
contains several buildings, which are the seminary buildings present today (Figure 2). In
1949, a footbridge was sited across the stream at the location of the present footbridge. It is
not known when the footbridge was constructed prior to 1949. A trackway leading south
from the footbridge follows a similar route as the current Fuk Luk Tsuen footpath.
- 72 -
In 1949? the course of the natural stream involved in the recent flooding appears to
follow much the same alignment as the present stream course. A minor tributary of the
stream was present between the seminary buildings and the area to the east of the trackway.
This tributary is present today as a dry gully feature, although it is no longer joined to the
valley stream (Figure 2).
From 1949 to the commencement of the construction of the estate platform in the late
1980's, the aerial photographs show a history of reducing cultivation at the estate site with a
corresponding increase in the development of small buildings.
The valley upstream of the estate shows relatively little change from the 1949 aerial
photographs to the present day. The more notable changes include the construction of the
WSD weir and associated shaft in the upper valley by 1975, and the Shing Mun Tunnel Road
viaduct across the lower valley by 1992 (Figure 1).
A review of documentary records, summarised in Appendix A, shows that the estate
site was part of planning area C4D' which was zoned for residential development, as shown on
the Shatin outline development plan issued in 1979 by the New Territories Development
Department (NTDD, renamed Territory Development Department in 1986). A proposed
layout for the area was subsequently issued in 1980 by NTDD.
In 1981, NTDD passed the proposed layout of area 4D to the Geotechnical Control
Office (GCO) amongst others for comment. In response, the GCO stated that "No objection
is raised against the proposed use of the above planning area". The GCO also included
comments on the geology and engineering of the site, and stated that "both main southward
draining tributaries of the Shing Mun River (which cross area 4D) have reasonably large
catchment areas and adequate drainage measures should be required as the first stage of
development. The photographs, particularly those of the 1960's, show severe sheet erosion of
the granite especially on the ridge crests and considerable siltation of drainage may therefore
be anticipated".
In 1983, a geotechnical study of area 4D by consultants Maunsell Consultants Asia
(MCA) was prepared for NTDD based on the 1980 proposed layout. The study recommended
an alternative layout, which subsequently formed the basis for the revised layout issued by
NTDD in 1984. In their comments on the MCA study report, the GCO stated that they had
"no adverse comments on the stability analysis, and the site formation layout is therefore
feasible provided the groundwater assumptions can be substantiated by MCA".
Further geotechnical reports for area 4D were produced by MCA, namely, a site
formation report in 1986, a drainage design report for the proposed slope drainage system in
1987, and supplementary geotechnical assessment reports in 1992 and 1993 that substantiated
the groundwater assumptions adopted in the 1983 report. The design of the slope drainage
system included consideration of surface run-off from the slope and adjacent related slopes,
but it did not include flow from the stream involved in the flooding incident. With respect to
the geotechnical design of the slopes and associated drainage, the GCO stated that they had
"no adverse comment in general on the proposed drainage system".
- 73 -
In 1989, in response to NTDD's proposal to use area 4D as a supplementary housing
site, the GCO stated that "there are no objections in principle to the use of the five areas
(including area 4D) as supplementary housing sites provided that the geotechnical limitations
affecting these sites are considered and dealt with in an appropriate manner during site
formation". The GCO's response also stated that "the site is located on terrain identified in
the Geotechnical Area Studies Programme (GCO, 1987) as colluvium and drainage plain",
and that "attention to both subsurface and surface drainage is needed to reduce the possibility
of slope instability and flooding".
The formation of the estate site based on MCA's design was completed by 1990. The
formed site included four cut platforms and the slopes on the estate boundary, including Slope
Nos. 7SW-D/C790, 7SW-D/C791 and 7SW-D/CR801. In 1991, the Housing Department
submitted a geotechnical report to the GCO which modified the site layout in order to
maximise the development potential. The report included the design for enlargement of
platforms and inclusion of a retaining wall. The report was checked by the GCO who stated
that they had "no adverse comments on the proposed site formation works to enlarge the
existing platforms".
Mei Chung Court platform remained vacant until the commencement of building
construction in 1995 which was substantially completed in early 1996.
In May 1997, Slope No. 7SW-D/C790 was inspected and registered in the New
Catalogue of Slopes as part of the GEO's project entitled 'Systematic Identification and
Registration of Slopes in the Territory' (SIRST) to systematically update the 1977/78
Catalogue of Slopes.
2.3
Land Status
The land status within the area of the flooding, as established by the District Lands
Office, is presented in Figure 2. Several private lots are present to the north of the estate,
associated with a seminary. However, the footpath and footbridge are located on Government
land.
3.
THE INCIDENT
3.1
Description of the Incident
The description of the incident is based on accounts from eye-witnesses at the estate,
from site observations by HAP and, in particular, an incident report by the Housing
Department (1997). A plan showing the details of the flooding incident is presented in
Figures.
During the rainstorm on 2 July 1997, stormwater flow within the natural stream course
over-spilled the stream banks at the location of a blockage at a footbridge
(Figure 3). Stormwater flooded onto a nearby footpath and, in part, was diverted downhill
towards Mei Chung Court, sited on a cut platform below the level of the footpath.
- 74 -
At about 05:25 hours whilst on patrol, the estate management superintendent observed
heavy rainwater run-off (floodwater) from the Fuk Lok Tsuen footpath spilling onto the
slopes behind Lok Chung House of Mei Chung Court (Figure 3). The floodwater entered the
estate at several locations causing severe erosion of the formed slopes.
Floodwater flowed through the estate causing erosion and undermining of paved areas,
notably within a play area to the east of Lok Chung House. The floodwater entered the lower
level of the multi-storey car park where it was impounded (Figure 3). Shortly after
05:25 hours floodwater had already entered the car park and was up to 0.6 m deep. The lower
level of the car park was reported by management staff to be inundated to a depth of 2.6 m by
05:48 hours. At 06:15 hours floodwater started to overflow from the car park and into the
lower floor of Fu Chung House. At 07:00 hours floodwater started to enter the meter room of
Lok Chung House.
At 05:48 hours the estate management staff reported the incident to the Police. The
Police and Fire Services Department (FSD) arrived on site at 05:55 hours. Divers from the
FSD commenced a search of the submerged lower level of the car park at about 09:00 hours.
No persons were found. At 10:40 hours the Police contacted the GEO for geotechnical advice
on the affected slopes. The GEO inspected the site at 11:45 hours (GEO Incident Report
No. MW97/7/7).
The flow of floodwater into the estate is reported to have continued for most of the
morning, with a notable reduction in flow after about 11:00 hours, which corresponds with an
amelioration in rainfall (see Section 4). Due to the potential for further heavy rain, at
17:00 hours Police evacuated the occupants of the seminary buildings located to the north of
the estate (Figure 3). Following an increase in rainfall between about 19:00 hours and
19:30 hours, flooding of the lower level of the car park was again reported.
Rain continued until about 04:00 hours on 3 July 1997 before diminishing, with no
further reports of serious flooding at the estate.
Two separate flood warnings were issued during the time of the incident, from
05:00 hours to 11:55 hours on 2 July 1997 and from 15:15 hours on the 2 July 1997 to
09:30 hours on 3 July 1997. A landslip warning was also in effect, from 06:25 hours on
2 July 1997 to 8:40 hours on 5 July 1997.
3.2
Consequential Damage
Flood damage within the estate included severe erosion of slopes concentrated at six
locations (Figure 3), undermining and erosion of paved areas mainly concentrated in the play
area to the east of Lok Chung House, damage to fixtures in Lok Chung and Fu Chung House
and damage to around seventy vehicles within the car park. The Fuk Luk Tsuen footpath
above the estate suffered gully erosion and undermining.
A photograph taken on 3 July 1997 is shown in Plate 1. This shows the accumulated
sediment in the lower level of the multi-storey car park after floodwater had subsided.
- 75 -
3.3
Immediate Follow-up Action
Following the incident, to prevent recurrence of flooding, the existing footbridge was
demolished, obstructions within the natural stream course removed, and a new bridge
constructed. The eroded section of Fuk Lok Tsuen footpath was also repaired.
The Drainage Services Department (DSD) is examining the long-term drainage
improvement to the area.
4.
RAINFALL
Analysis of rainfall data recorded at the time of the flooding has been carried out to
determine the characteristics and estimated return period of the rainfall event. The rainfall
data from several GEO automatic raingauges (Nos. N01, N02, N06 and N14) in the vicinity of
the site were reviewed. The raingauges record and transmit rainfall data to the GEO and the
Hong Kong Observatory (HKO) at 5-minute intervals. The rainfall data from individual
raingauges showed similar characteristics, and therefore the nearest raingauge to Mei Chung
Court, raingauge No. N02, was selected as representative of rainfall at the site.
The daily and hourly rainfall recorded during June and July 1997 are shown in
Figure 4. The daily rainfall during the 31-day period preceding the flooding incident was
generally below 20 mm, and only exceeded 100 mm on 4 June and 16 June 1997. The hourly
rainfall record during the days preceding the incident showed generally no rainfall, except for
a rainstorm event from 04:00 hours to 12:00 hours on 1 July 1997, with a maximum rolling
hourly intensity of 27 mm/hr. The 24-hour rainfall total prior to the incident was about
120mm.
The incident occurred during the early stages of a rainstorm event that commenced at
about 04:00 hours on 2 July 1997. By the time of the onset of flooding, at 05:25 hours, the
site had received between 40 mm and 60 mm of rainfall as shown in Figure 5.
The rolling 60-minute rainfall immediately preceding the incident was 43 mm/hr. An
analysis of the return period for this rainfall event using the historical rainfall records at the
HKO (Lam & Leung, 1994) indicates that the estimated return period is 1 year. The rainfall
during the preceding 31-day period indicates that this 60-minute rolling rainfall total was
previously exceeded during the rainstorm of 4 June 1997, with 63 mm/hr.
Rainfall intensity increased after the onset of flooding (Figure 6) and greatly escalated
the scale of the incident. The maximum 60-minute rolling rainfall intensity for the rainstorm
was 124 mm/hr, which occurred between 05:25 hours and 06:25 hours on 2 July 1997, with
an estimated return period of about 32 years.
- 76 -
5.
SOURCES OF FLOODWATER
From field observations, the major source of floodwater was found to be stream flow
diverted from the natural stream course at the footbridge. Stormwater would have originated
as run-off from the catchment shown in Figure 1.
The possibility of an increased flow in the natural stream due to overflow of the WSD
weir located at the upstream end of the catchment was investigated. The WSD commented
that "the intake was designed to abstract stream water from upstream. It is provided with a
spillway for overflow of stream water in the case that the level of the stream rises to the
spillway level. It should be noted that the quantity of water overflow would always be the
natural yield deducted by the water abstracted by the intake (shaft) and be less than the natural
flow. Therefore there was no increased flow in the natural stream due to overflow of the
WSD weir. It can be viewed that the natural storm is attenuated by the water abstraction
under such circumstances."
The likely stormwater flow at the footbridge and response time to rainfall was
determined from a synthesised hydrograph for the catchment area, assuming no over-spill of
the WSD weir (Water Authority, 1968). The results, which are considered indicative, are
shown with the 5-minute rainfall record in Figure 6. The response time between rainfall and
peak stormwater flow is estimated at 14 minutes, with an estimated peak stormwater flow of
1.7m3/sec.
The discharge capacity of the stream channel beneath the footbridge was compared
with the indicative stormwater flow at 05:25 hours (Figure 6). A scoping calculation indicates
that about an 80% reduction in channel cross-section was necessary to prevent stormwater
flow from passing beneath the footbridge. If the channel was unobstructed, there was
sufficient capacity beneath the bridge to accommodate the theoretical peak flow of
1.7mVsec.
Other sources of floodwater, such as run-off from the cut slope (No. 7SW-D/C787) to
the west of the footpath (Figure 2), are considered minor in comparison with the amount of
stormwater from the stream.
6
-
6.1
CONDITIONS AND DAMAGE AT THE ESTATE AND SURROUNDING AREA
General
A walkover inspection was undertaken to establish conditions within the natural
stream course, the surrounding area and the estate following the flooding. These are detailed
below and presented in Figure 3.
6.2
Natural Stream Course
On the upstream side of the footbridge an accumulation of debris (comprising silty
sandy gravel with many sub-rounded cobbles and some sub-rounded boulders) was found to
- 77 -
have completely blocked the upstream side of the bridge. This can be seen in a photograph
taken during the morning of 3 July 1997 (Plate 3).
Clearance of debris from the channel upstream of the bridge by the DSD revealed up
to about 0.8 m depth of generally loose debris underlain by more bouldery debris (Plate 4).
The space beneath the bridge, on the unblocked downstream side, was about 1.7 m wide by
1 m high (Plate 5). The width of the stream course either side of the bridge varied, but
typically ranged from 3 m to 4 m.
The deck of the footbridge was constructed in concrete (Plate 5), the appearance of
which suggested the possibility of a later date of construction than 1949 (the earliest available
aerial photographs, see Section 2.2). The construction of the deck may coincide with the
resurfacing of the footpath, as identified in the 1992 aerial photographs (Table 1).
Based on site observations, the debris removed from the stream course appeared to be
consistent with transported stream deposits. It was not possible to determine the length of
time the debris had been deposited at this location.
An inspection of the natural stream course further upstream of the footbridge did not
reveal any significant evidence of a concentrated source of debris that may have entered the
stream and contributed to the blockage at the footbridge. Only a minor, possibly recent
landslide of about 20 m3 in volume was identified on the stream bank, some 350m upstream
of the footbridge (Plate 6). It is not considered likely that this landslide contributed
significantly to the debris at the footbridge.
The most likely source of debris contributing to the blockage was the stream bed
deposits. The deposits comprised transported detrital material, grading from silty sandy
rounded gravel to rounded to sub-rounded cobbles and boulders.
6.3
Area between the Estate and Footbridge
Uphill of the estate, erosion gullies formed parallel to the footpath from the footbridge
to the estate, exploiting the existing surface drainage channel and an abandoned stream course
(Figure 3). A photograph taken on 3 July 1997 shows the extent of the erosion (Plate 7).
Where floodwater entered the estate, flood debris accumulated against the estate boundary
fence (Plate 8).
6.4
The Estate
The floodwaters which reached the estate caused serious erosion to two cut slopes
(Nos. 7SW-D/C791 and 7SW-D/CR801), to the north of the multi-storey car park and Slope
No. 7SW-D/C790 to the north of Lok Chung House (Figures 2 and 3).
Erosion scars were observed on the cut slope (No. 7SW-D/C791) at four locations
beneath the section where floodwater entered the estate or over-topped the slope drainage
(Plate 9). The scars were generally less than 1 m deep but locally up to 2 m deep. Exposures
- 78 -
in the erosion scars showed localised fill up to 1 in deep overlying completely decomposed
granite. It is estimated that a total volume of about 45 m3 of soil was washed from the slope.
Erosion scars were concentrated at two locations on Slope No. 7SW-D/C790. The
scars to the northwest of Lok Chung House were shallow, typically 0,5 m deep within
completely decomposed granite. To the northeast of Lok Chung House a further scar, over
2 m deep exposed alluvial deposits (of dense sandy gravel with many sub-rounded cobbles
and occasional sub-rounded boulders) overlying completely decomposed granite (Plate 10).
The estimated volume of the soil removed from the slope was 55 m .
Paved areas beneath the slopes, including the play area, were undermined as
floodwater followed the general fall towards the multi-storey car park. The ground floor of
Lok Chung and Fu Chung House were also affected by floodwater causing damage to fixtures
and lifts.
Staining on the walls and columns inside the basement level of the car park indicated a
floodwater level of about 1.7 m (Plate 2), though some eye-witnesses stated that the flooding
was up to 2.5 m deep. The floodwater deposited soft sandy silt debris in the basement, with
an estimated volume of about 100 m3.
7.
PROBABLE CAUSES OF THE FLOODING INCIDENT
The serious erosion of the slopes and flood damage at Mei Chung Court on
2 July 1997 were the result of diverted storaiwater flow from a natural stream course which
had become blocked with debris. The following factors probably contributed to the incident:
(a) the presence of a natural stream course with the potential
for increased mobilisation of stream deposits during storm
flow conditions,
(b) the presence of a local constriction In the natural stream
course at a footbridge resulting in accumulation of stream
deposits and consequent blockage of the stream, and
(c) the presence of a preferential surface drainage path, namely
the footpath and a minor abandoned tributary stream course
(Figure 3), which allowed water to flow towards the estate
and over-spill the crest of the slopes on the northern
boundary of the estate.
8.
CONCLUSIONS
The flooding was initiated during the early morning (at about 05:25 hours) on
2 July 1997 when the preceding 60-minute rolling rainfall was 43 mm/hr with an estimated
return period of only about one year. Therefore, it is concluded that the channel was probably
significantly obstructed prior to the onset of the flooding.
- 79 -
The scale of the flooding greatly escalated as the rainfall intensity increased to a peak
intensity of 124 mm/hr between 05:25 hours and 06:25 hours.
The affected slopes, designed to current geotechnical standards, did not have any
overall slope instability problems despite being inundated by floodwater and debris during the
incident. However, the surface water drainage and slope protection measures were not
designed to take account of the possibility of over-spill from the natural stream course, and
consequently severe surface erosion of material from the slopes occurred during the flooding
incident.
9.
REFERENCES
Territories. Geotechnical Control Office. Hong Kong, GASP Report no. II, 165 p., plus
4 maps.
Housing Department (1997). Incident Report Landslip and Flooding in Mei Chung Court
(MCC). Memorandum from HM/Mei Chung Court to CM/M (NTE), reference
HD(H)CNS 11/22/1, 8 July 1997.
selected locations in Hong Kong. Royal Observatory Hong Kong. Technical Note
no. 86, 89 p.
Water Authority (1968). Design Flood for Hong Kong. Water Authority, Public Works
Department, Hong Kong Government, 49 p.
- 80 -
LIST OF TABLES
Table
No.
1
Page
No.
Summary of Site Development from API
81
- 81 -
Table 1 - Summary of Site Development from API
Year
1949
Photographic
Reference No.
Y02112,
Y02113
Attitude
(feet)
5800 ft
1963
5349, 5350
1975
11553, 11554
3000 ft
1977
20078, 20079
4000 ft
1984
56850,56851
4000 ft
1986
A04906,
A04907
4000 ft
1992
A3 1920,
A31921
4000ft
1995
CN10822,
CN10823
CN14957,
CN14953
2500 ft
1996
4000ft
Observation
Seminary buildings (now the Asian Christian Conference
Centre) are present north of Mei Chung Court (MCC) site,
including trackway and footbridge at the location of the
1997 flooding incident. The majority of the MCC site is
cultivated land, irrigated fields fed by a minor tributary
stream are present in the northeast. The tributary stream runs
from the main valley stream to the west of the seminary
buildings. The seminary is situated on high ground between
the tributary and main valley streams. Boulders visible in
main stream at location of footbridge.
Field network increased and more buildings present within
valley to north of MCC site. Footbridge is still present. The
trackway leading to the footbridge has increased in width.
Irrigated fields are still present in northeast of MCC site.
Another footbridge is constructed south of seminary.
Large scale development of small buildings on fields in
south of MCC site. Irrigated fields are still being worked in
northeast of MCC site.
Another footbridge built
downstream to east of the school. Weir structure and
overflow vertical shaft constructed in the mid-upper section
of the valley.
Small building development continuing on MCC site.
Cultivation of fields on MCC site in decline. Tree growth
obscures the view of the footbridge.
Mei Shing Court is under construction. The valley stream is
cul verted south of MCC site.
No significant new
developments upstream. Change of cultivation type in fields
in northeast of MCC site, appear not to be irrigated.
Construction of Mei Shing Court is completed. Footbridge is
not clearly shown. Cultivation of fields in northeast of MCC
site in decline.
Platform area and slopes at the MCC site are completed.
Footpath connecting to the footbridge has a high reflectance
suggesting it was recently re-surfaced. Slopes along the west
of the trackway formed and stream trained at the location of
viaduct. Boulders within the stream bed are clearly visible
in vicinity of footbridge. Shing Mun Tunnel Road is
completed.
MCC is under construction. Boulders present at the
upstream edge of footbridge.
Construction of Mei Chung Court is substantially completed.
- 82 -
LIST OF FIGURES
Figure
No.
Page
No.
1
Site Location Plan
83
2
Plan of the Flood Affected Areas
84
3
Plan of Area Showing Details of Flooding Incident
85
4
86
5
Isohyets of Rainfall between 03:50 Hours and 05:25 Hours
on 2 My 1997
87
6
Rainfall Recorded at GEO Raingauge N02 at 5-minute Intervals
and Stormwater Flow in Natural Stream at Footbridge
88
-83 -
^J
\
New TerTitories < S_ r-fC
The site
V^7
"
Key Plan
WSD weir and
drainage shaft
Drainage catchment boundary
of natural stream course from
WSD weir to footbridge
j
j
S27MOJ4
0 20 40 60 m
J
L.
Scale
Note :
Base map is extracted from Survey Sheet Nos. 7SW-14B, 7SW.-14D, 7SW-15A
and 7SW-I5C dated October 1992 (Original scale 1:1 000).
-84
DO
03
CD
0)
4~»
Q
f<D
O
J2
E
CD
4»a
CM
O
CL,
CNS
0)
3
t£
-85 -
a
*i-H
T3
O
o
o
Cfl
'3
•4-t
Q
o
^
C/}
CS
a>
O
.SP
£
- 86 -
700
Time of flooding
(beginning at 05:25 hours)
----
600 4
-
-
i
•
T '
500
If
^
1
400 4
300 -
200 4
1
n
2
h I.,
4
6
8
10
12
14
16
18
i.i.i ..IlL
20
22
24
26
28
2
30
June 1997
(a) Daily Rainfall Recorded between 2 June and 5 July 1997
122 mm of rainfall recorded in the
24 hours before the flooding
120 .
Time of flooding
(beginning at 05:25 hours)
100 4
80 &
'I
60
I
4
°
20 4
0
0:00
•ll.-l -
JUL
6:00
12:00 18:00
30 June 1997
0:00
6:00
1 July 1997
klh
_L
12:00 18:00 0:00
6:00
4
July 1997
12:00 18:00
2 July 1997
(b) Hourly Rainfall Recorded between 30 June and 2 July 1997
87-
ON
OS
f—H
3
o
en
tH
o
<N
in
O
T3
a
cd
CO
5-H
^
O
ffi
o
CO
O
ri
C4H
O
«/3
•4-*
a>
^
43
o
en
- 88
Ul)
_
o
o
PH
I
•4—>
00
I
2
J2
E»-i
Is
o
GO
13
CD
I—H
I
CM
O
2
o
2
CD
W)
c3
bD
O
w
o
•f->
cd
T3
CD
•2O
o
*s
(^
i
vo
I
bD
- 89 -
LIST OF PLATES
Plate
No.
4
Page
No.
1
View of Basement Level (L1) of Multi-storey Car Park
90
2
View of Floodwater Staining, up to 1.7 m, in Car Park
(Photograph Taken in August 1997)
90
3
View of Footbridge where Stream Over-spilled its Channel
91
4
General View of Upstream Edge of Footbridge Following
Clearance Works (Photograph Taken in August 1997)
91
5
General View of Unblocked Downstream Edge of Footbridge
(Photograph in August 1997)
92
6
View of Possible Recent Landslide on Stream Bank
92
7
Gully Erosion alongside Fuk Lok Tsuen Footpath
93
8
Accumulation of Flood Debris against Estate Boundary Fence
93
9
Photograph Showing Flood Damage to Cut Slope
No. 7SW-D/C791 at Mei Chung Court (Photograph Taken
on 3 July 1997)
94
10
View of Erosion Scar and Undermined U-Channel on Slope
No. 7SW-D/C790 (Photograph Taken on 3 July 1997)
95
- 90 -
Plate 1 - View of Basement Level (LI) of Multi-storey Car Park
Plate 2 - View of Floodwater
Staining, up to 1.7m,
in Car Park
(Photograph Taken in
August 1997)
- 91 -
Plate 3 - View of Footbridge where Stream Over-spilled its Channel
Plate 4 - General View of
Upstream Edge of
Footbridge Following
Clearance Works
(Photograph Taken in
August 1997)
- 92 -
Plate 5 - General View of Unblocked Downstream Edge of Footbridge
Plate 6 - View of Possible
Recent Landslide on
Stream Bank
(Photograph in August
1997)
- 93 -
Plate 7 - Gujly Erosion alongside Fuk Lok Tsuen Footpath
Plate 8 - Accumulation of Flood Debris against Estate Boundary Fence
94-
- 95 -
Plate 10 - View of Erosion Scar and Undermined U-Channel on Slope
No. 7SW-D/C790 (Photograph Taken on 3 July 1997)
- 96 -
APPEM)IX A
SUMMARY OF DOCUMENTS REVIEWED
- 97 -
Table Al - Summary of Documents Reviewed (Sheet 1 of 6)
Document Title
By Whom
(Draft) Shatin
outline Development
Plan No. 77/202BD
NTDD
To
Whom
Various
Layout Plan STP
80/008 for Area 4D
NTDD
Various
Shatin New Town
Development Sketch
Plan No. STP80/008
Planning Area 4D
Shatin
GCO
NTDD
Remarks
Mel Chung Estate site located
within planning area 4D.
Zoned for private residential
development.
Sketch plan of proposed layout
of planning area 4D.
Geotechnical comments on
sketch plan No. STP80/008. In
their comments, the GCO stated
"No objection Is raised against
the proposed use of the above
planning area". Comments on
the geological and engineering
aspects of area 4D were also
included. With particular
reference to drainage, the GCO
commented that, "Both main
southward drainage tributaries
of the Shing Mun River have
reasonably large catchment
areas and adequate drainage
measures should be required as
the first stage of development.
The photographs, particularly
those of the 1960's, show
severe sheet erosion of the
granite especially on the ridge
crests and considerable siltation
of drainage may therefore be
anticipated."
- 98 -
To
Whom
GCO
Remarks
Year
Document Title
By Whom
1981
Geotechnical
Checking of New
Works Shatin New
Town Development
- Planning Area 4D
Binnie and
Partners
(Hong
Kong)
1983
Shatin New Town
Area 4D Site
Formation Proposals
Stability Analysis of
Slopes
Maunsell
Consultants
Asia
(MCA)
NTDD& Study report examined stability
analysis of existing slopes.
GCO
Alternative layout
recommended.
1984
Shatin New Town
Formation, Roads &
Drains in Area 4D
Geotechnical Study
- Proposed Platform
Layout
GCO
NTDD
In response to a request by the
GCO, Binnie and Partners
(B&P) visited and undertook an
API of area 4D to assess the
stability of slopes.
B&P commented that, "detailed
investigation is required to
establish the existing stability
and hence the extent of
development acceptable, An
assessment of groundwater
levels, based on piezometer
readings will be of particular
importance."
Review of MCA (1983) study
report. GCO comments that,
"The study report delineates a
revised site formation layout
that satisfies the safety factor
requirements, and the design of
the slopes is backed up by slope
stability analysis". The GCO
concluded that they "have no
adverse comments on the
stability analysis, and the site
formation layout is therefore
feasible provided the ground
water assumptions can be
substantiated by MCA".
- 99 -
Year
Document Title
By Whom
1984
Shatln New Town
Development
Planning Area 4D,
Layout Plan No.
STP/L4D/5
Explanatory Notes
NTDD
1986
Shatin New Town
Route 5 - Shatin
Connection Phase 1
Area 4D Site
Formation
Geotechnical
Submission
MCA
TDD *
&GCO
Report examined and analysed
the stability of slopes adjacent
to, and between platforms for
the proposed development.
Also included an assessment of
ground water levels at critical
sections.
1986
(June
and
Oct)
Shatin New Town
Formation, Roads &
Drains in Area 4D
Geotechnical
Submission
GCO
TDD
Checking of MCA (1983)
report found acceptable
however comments that final
assessment on groundwater
monitoring is outstanding.
1987
Shatin New Town,
Stage II Contract
No. ST24/85 Route
5 - Sha Tin
Connection Phase 1
Drainage Design
MCA
GCO
Proposed slope drainage system
design submission included
design calculations and
drainage layout plans. Plans
show drainage design allows for
surface run-off from cut slopes
from nearby related slopes.
1987
Shatin New Town,
Stage II Contract
No. ST24/85 Route
5 - Shatin
Connection Phase 1
Drainage Design
GCO
TDD
In response to the slope
drainage design submitted by
MCA (1987), the GCO stated
that they had "no adverse
comment in general on the
proposed drainage system".
To
Whom
Various
Remarks
Revised site layout plan based
on MCA (1983) and
accompanying explanatory
notes.
* NTDD renamed Territory Development Department in 1986
- 100 -
Year
Document Title
By Whom
1989
Supplementary
Housing Sites in
Shatin
GCO
1991
Shatin Area 4D
Supplementary Site
Formation
Geotechnical Report
Civil
Engineering
Section
Construction
Branch, HD
To
Whom
TDD
GCO
Remarks
GCO provided geotechnical
comment on proposed housing
sites in the Shatin area,
including planning area 4D. The
GCO stated that, "There are no
objections in principle to the
use of the five areas [including
area 4D] as supplementary
housing sites provided that
geotechnical limitations
affecting these sites are
considered and dealt with in an
appropriate manner during site
formation." The GCO went on
to state that, "This site [area
4D] is located on terrain
identified in the Geotechnical
Area Studies Programme as
colluvium and drainage plain.
It lies across two natural
drainage lines with a steep and
extensive catchment area", and
that, "Attention to both
subsurface and surface drainage
is needed to reduce the
possibility of slope instability
and flooding."
Stability analysis of slopes and
retaining walls modified from
the original site layout (MCA,
1983) to optimise development
potential. Modification
included enlargement of
platforms creating a fill slope
and retaining wall within the
site and increase (of 1 m) in cut
slopes in northwest of site.
- 101 -
Year
Document Title
By Whom
1991
Shatin Area 4D
Supplementary Site
Formation
Geoteclmical Report
GCO
1992
(Oct)
Shatin New Town
Stage II Contract
No. ST24/8S Route
5 - Shatin
Connection, Phase 1
Area 4D
Supplementary
Geotechnical
Assessment
MCA
HDand
GEO
Review of additional
geotechnical information for the
site, included piezometer
records, triaxial test results and
borehole logs.
1992
(Nov)
and
1993
(Mar)
Shatin New Town
Stage II Contract
No. ST24/85 Route
5 - Shatin
Connection, Phase 1
Area 4D
Supplementary
Information
MCA
HDand
GEO
Groundwater assessment of site
slopes. MCA found "that the
groundwater table is not critical
and will not affect the site".
1993
Home Ownership
Scheme STTL396,
Shatin Area 4D
GEO
In response to MCA (1993),
HD,
DLO and GEO had "no further comment"
with regard to groundwater
NENT
table at the site.
1996
(Sept)
Phase 2 Systematic
Inspection of
Features in the
Territory (SIFT 2
Version 4.4)
Consultants
to GEO
GEO
To
Whom
HD
Remarks
In response to the HD (1991)
geotechnical report the GCO
stated they had "no adverse
geotechnical comments on the
proposed site formation works
to enlarge the existing
platforms."
Identification of slope (7SWD/C790) from API.
- 102 -
Year
Document Title
By Whom
1997
(May)
SIRST- Field Sheet Consultants
and Previous History toGEO
/ Documentation Background
Information
To
Whom
GEO
Remarks
Provided details of slope No.
7SW-D/C790, including field
observations and background
information.
- 103 -
SECTION 4 :
YAU KOM TAU
ON 2 AND 3 JULY 1997
This report was originally produced in May 1998
- 104 -
FOREWORD
flooding incident on 2 and 3 My 1997, which caused considerable
disruption to the east- and west-bound carriageways of Tuen Mun
Road. Stormwater from the Tai Lam Chung Catchwater overflowed
into a natural stream course as a result of blockage of the catchwater
with debris from three landslides.
The stormwater caused
considerable erosion of the stream bed and banks. Debris from the
stream was carried into and choked the drainage system above a
registered cut slope adjacent to Tuen Mun Road. As a consequence,
water and debris was discharged onto the road and caused its closure
for 5.5 hours west-bound and up to 15.5 hours east-bound. No
fatalities or injuries were reported.
facts about the incident, present relevant background information and
establish the probable causes of the incident. The scope of the study
report was written by Mr R J Simonds and reviewed by Dr R Moore
and Mr H Siddle. The assistance of the GEO in the preparation of
the report is gratefully acknowledged.
G. Daughton
- 105 -
CONTENTS
Page
No.
Title Page
103
FOREWORD
104
CONTENTS
105
1.
INTRODUCTION
106
2.
THE SITE
107
2.1
Site Description
107
2.2
Site History
107
2.3
Previous Studies
108
2.3.1
Slope Registrations
108
2.3.2
Inspections of Catchwater Cut Slope No. 6SE-D/CR135
109
2.3.3
Inspections and Stage 1 Study of Cut Slope No. 6SE-D/C64
109
2.3.4
Catchwater Studies and Subsequent Actions
110
3.
2.4
Previous Landslides
111
2.5
111
THE INCIDENT
112
3.1
The 1997 Landslides
112
3.2
The Flooding Incident
112
4.
RAINFALL
113
5.
CAUSES OF THE LANDSLIDES AND FLOODING INCIDENT
114
6.
CONCLUSIONS
116
7.
REFERENCES
116
LIST OF TABLES
118
LIST OF FIGURES
125
LIST OF PLATES
135
- 106 -
1.
INTRODUCTION
Prior to midnight on 2 July 1997 three landslides occunred on cut slope
No. 6SE-D/CR135 above the Tai Lam Chung Catchwater near Yau Kom Tau new village
resulting in complete blockage of the catchwater (Figures 1 and 2). As a consequence,
stormwater was discharged through an overflow weir 'X' (Figure 2) into a natural stream
course below. The stormwater eroded the banks and bed of the stream including dilapidated
masonry walls constructed on abandoned squatter platforms adjacent to the stream. Boulders
and debris mobilised by the stormwater blocked a trapezoidal drainage channel located at the
crest of cut slope No. 6SE-D/C64 above the east-bound carriageway of Tuen Mun Road. As a
result of the blockage of the trapezoidal drainage channel, stormwater and debris cascaded
down the cut slope and the intervening fill slope (No. 6SE-D/F17) onto the east- and westbound carriageways of Tuen Mun Road, causing flooding and total blockage of the road. The
east- and west-bound carriageways were closed for 15.5 hours and 5.5 hours respectively. No
fatalities or injuries were reported.
Following the flooding incident, Halcrow Asia Partnership Ltd (the 1997 Landslip
Investigation Consultants) carried out a detailed study of the incident for the Geotechnical
No. CE 68/96. This is one of a series of reports produced during the consultancy by Halcrow
The key objectives of the study were to document the facts about the incident, present
relevant background information and establish the probable causes of the incident. The scope
of the study was limited to site reconnaissance, desk study and analysis. Recommendations
for follow-up actions are reported separately.
key tasks:
site,
(c) detailed observations and measurements at the site,
including a height survey of the catchwater parapet wall
and overflow weirs,
(d) a review of reports produced after the incident by
Highways Department (HyD) and Water Supplies
Department (WSD), and
(e) diagnosis of the probable causes of the incident.
- 107 -
2.
THE SITE
2.1
Site Description
The flooding incident occurred on Tuen Mun Road below the abandoned squatter
village of Hon Man Tsuen, about 100 m east of the new village at Yau Kom Tau (Figure 1).
Tuen Mun Road is grade-separated at the location of the flooding incident, with the eastbound carriageway some 7 m above the west-bound carriageway. The carriageways are
separated by fill slope No. 6SE-D/F17. A partly modified natural hillside is present above the
7 m high cut slope (No. 6SE-D/C64) adjacent to the east-bound carriageway.
The three landslides occurred on cut slope No. 6SE-D/CR135 above the Tai Lam
Chung Catchwater at locations C A', C B' and C C (Figure 1 and Plates 1 to 3). The slope is
about 550 m long and has been cut at an angle of about 60° with no berms, into soil and rock
to a maximum height of about 18 m. About 40% of the slope face is sealed, 50% vegetated
and 10% remains bare. The types of seal along the slope face vary between shotcrete,
chunam and stone pitching. There is a 250 mm U-channel crest drainage system which
connects to a number of 300 mm stepped channels which discharge into the catchwater.
Running along the length of the toe of the slope is a 1.7 m high concrete retaining wall which
forms the upslope side of the catchwater.
The hillside between the Tai Lam Chung Catchwater and Tuen Mun Road is about 23°
to the horizontal, and has in the past been modified to form cut slopes and platforms for
squatter huts. The heavily vegetated natural hillside above the catchwater cut slope forms the
southern flank of Ha Fa Shan and is at an angle of about 34° to the horizontal
The Tai Lam Chung Catchwater is a concrete-lined channel approximately 1.6 m deep
and 2.2 m wide. There is a 3 m wide WSD access road adjacent to the catchwater separated
by a concrete parapet wall which is about 0.5 m high. Overflow weirs on the catchwater
regulate water flow within the catchwater. One such overflow weir C X' (Figure 1) regulates
water flow into a natural stream course which passes through the abandoned squatter village
of Hon Man Tsuen. The stream discharges into a trapezoidal drainage channel C D' (Figure 1),
which is about 0.8 m deep and 1.5 m wide. A flat concrete slab forms an access footbridge
over the outfall of the trapezoidal drainage channel, which discharges into a cascade channel
C
E' (Figure 1) down cut slope No. 6SE-D/C64. A 1.5 m diameter culvert T' (Figure 1) is
located at the bottom of the cascade channel allowing water to drain beneath Tuen Mun Road.
As part of the detailed study, HAP carried out a height survey of the catchwater
parapet wall between overflow weirs C W, 'X', *Y' and C Z'. The results of the survey are
presented in Section 5.
2.2
Site History
The site history was traced from a sequential series of aerial photographs of the site
spanning the period 1954 to 1996 and a review of other available documentary information
(Tables 1 and 2).
Aerial photographs taken in 1954 show that prior to the construction of the Tai Lam
Chung Catchwater there were several prominent drainage lines on the natural hillside at Ha Fa
- 108 -
Shan. The section of catchwater affected by landslides was constructed in 1956 by WSD.
Aerial photographs taken in 1963 show the catchwater and cut slope and a heavily vegetated
natural hillside directly upslope. Bouldery colluvium can be seen in the stream bed below
overflow weir 'X' (Figure 3). A small number of squatter huts were present at Hon Man
Tsuen, having been constructed on platforms cut into the natural hillside below the catchwater
to the east and west of the stream.
The number of squatter huts in Hon Man Tsuen had increased by 1974. There were no
apparent changes to the Tai Lam Chung Catchwater cut slope, overflow weirs or the natural
hillside above the catchwater. Aerial photographs taken in 1975 show a colluvial debris fan
in the area through which the trapezoidal drainage channel 'D' was subsequently constructed.
The 1975 aerial photographs also show the U-channel crest drainage system above the
catchwater cut slope. The crest drainage system is connected to a number of stepped channels
down the catchwater cut slope. At this time the natural hillside above the catchwater was
heavily vegetated with the exception of a small area, partially bare of vegetation, which is
considered to be the scarp of a minor landslide (Figure 3). Aerial photographs show that Tuen
Mun Road and Slope Nos. 6SE-D/C64 and 6SE-D/F17 were constructed in two stages
between 1975 and 1984. Apart from a small amount of erosion in the area of the 1975
landslide, that is visible on photographs taken in 1976, there were no significant changes to
the Tai Lam Chung Catchwater or roadside slopes between 1984 and 1990.
The 1992 aerial photographs indicate that there may have been uncontrolled discharge
of surface water adjacent to cascade channel C E' on cut slope No. 6SE-D/C64. About one
year later, the dark surface staining was no longer present.
By 1993, some of the squatter huts in the village at Hon Man Tsuen had been rendered
uninhabitable and only the foundation slabs remained. These actions were the result of the
Non-Development Clearance (NDC) programme at Hon Man Tsuen which was implemented
jointly by Lands Department and Housing Department in 1992. The approximate area
covered by the NDC programme is shown in Figure 1.
2.3
Previous Studies
2.3.1 Slope Registrations
The catchwater cut slope No. 6SE-D/C135 and the highway cut slope No. 6SE-D/C64
were both registered by consultants engaged by the Government to prepare the 1977/78
Catalogue of Slopes.
In 1992, the GEO initiated a consultancy agreement entitled 'Systematic Inspection of
Features in the Territory' (SIFT) to update information on existing registered slopes based on
studies of aerial photographs. Slope Nos. 6SE-D/C135 and 6SE-D/C64 were both entered in
the SIFT database in July 1996, the former being allocated the new Slope No. 6SE-C/CR135
in recognition of the fact that the catchwater formed a retaining structure at the toe of the
slope. The category 'Class Cl' was assigned to both slopes as they had "been formed or
substantially modified before 30.6.78" and satisfied the criteria for slope registration.
- 109 -
2.3.2 Inspections of Catchwater Cut Slope No. 6SE-D/CR135
The catchwater cut slope No. 6SE-D/CR135 was inspected In December 1996 by
consultants engaged by the Government for the 'Systematic Identification and Registration of
Slopes in the Territory' (SIRST) to update the 1977/78 Catalogue of Slopes and to compile
the New Catalogue of Slopes. The condition of the cut slope face was recorded as being
"fair" with past instability referred to as "multiple minor". The stepped channels were
recorded as "clear"; however, the crest U-channel was reported as "blocked95. The concrete
retaining wall showed no "inferred past instability" though a U-charmel at the toe was
"blocked" and weepholes were "partially blocked". A further study of the slope was
recommended and this has been instigated by WSD (Section 2.3.4).
It is understood from WSD "that the slopes along the catchwaters are usually inspected
by our catchwater gang. The catchwater gang will report to their Inspector and Engineer the
abnormal features of the slopes for follow-up actions on the routine maintenance and repair
works of the slopes" (WSD, 1998a). Aerial photographs show evidence that a protective
surfacing was applied to cut slope No. 6SE-D/CR135 to the west of overflow weir 'X' at
sometime between 1963 and 1975 and near overflow weir 'X' at sometime between 1992 and
1993. WSD (1997a) reported that slope protection works were undertaken at three locations
on the catchwater cut slope in 1995 (Figure 3).
2.3.3 Inspections and Stage 1 Study of Cut Slope No. 6SE-D/C64
A Stage 1 Study of highway cut slope No. 6SE-D/C64 was undertaken in May 1989 by
the Geotechnical Control Office (GCO, 1989). The report recommended "no further study is
considered necessary for the slope under the present circumstances. However, based on site
inspection, the following maintenance works are considered necessary for the slope :
(a) clear the unplanned vegetation from the slope face and
repair the cracks on the chunam slope cover, and
(b) clear the drainage channels and repair cracks, if any."
In February 1996, the highway cut slope No. 6SE-D/C64 was inspected by consultants
Fugro Mouchel Rendel (FMR? 1995) appointed by HyD in a project entitled 'Roadside Slope
Inventory and Inspections' to, inter alia, carry out Engineer Inspections on about 4 000 HyD
slopes which satisfy GEO's slope registration criteria. The Engineer Inspection Report noted
that routine maintenance had not been carried out satisfactorily. In particular, culverts,
drainage channels and natural drainage lines were noted to be "blocked with rubbish". A
"high consequence to life" category was assigned to the slope and, together with yearly
inspections, a "detailed stability analysis of the suspect slope" was recommended. ^ The
Engineer Inspection Report stated that the overall state of maintenance of the slope was "fair"
according to guidelines given in the Guide to Slope Maintenance (GEO, 1995). A Stage 1
Study was subsequently carried out by FMR in February 1996 which recommended the need
for further study of Slope No. 6SE-D/C64.
- 110 -
2.3.4 Catchwater Studies and Subsequent Actions
In November 1979, WSD and GCO jointly commenced a study of catchwaters in
Hong Kong. The purpose of the study "was to investigate the stability of catchwaters and
associated slopes and to make appropriate recommendations to reduce the potential risk to
downslope developments." The Catchwater Study First Report (WSD & GCO? 1980)
provided recommendations for a systematic study and ranking of all catchwaters. These
recommendations were fulfilled by the Catchwater Study Second Report (GCO, 1982). The
principal conclusions were that "there are a small number of catchwaters which are of concern
for their whole length" and "there are a number of isolated sections of other catchwaters
which are of concern." Further studies were carried out "on six catchwaters considered to
present the greatest potential risks in terms of consequences of failure." One section of the
Tai Lam Chung Catchwater, Section 0, which includes the slope on which 1997 landslides
occurred was included in the study. The report (GCO, 1982) stated that "in the section
CH6465-7350 there are several squatter villages situated on the steep slopes between the
catchwater and the Tuen Mun Highway". Risk to the cut slope No. 6SE-D/C64 was also
noted. For all the catchwater sections of concern, the report recommended :
(a) "WSD undertake a detailed hydrological study,
"
(b) "A review of all recent remedial work carried out during
regular or special maintenance should be done
"
(c) "For the lengths of major catchwaters that remain at risk by
blockage or collapse WSD should then make specific
requests to GCO for further studies
"
In March 1986, WSD prepared a report (WSD, 1986) on the Tai Lam Chung
Catchwater addressing recommendations (a) and (b). The report considered the hydraulics of
the catchwater in relation to flood flows and compared a 1 in 200 year flood flow in the
catchwater with the existing catchwater overflow capacity. The report determined that
overflow weirs C W and 'X' were of sufficient design, however modifications were
recommended to increase the capacity of overflow weirs 'Y' and 'Z'. According to WSD,
"this section of the Tai Lam Chung Catchwater has been upgraded since 1992 to ensure that it
will cope with a risk of I to 200 year return period" (WSD, 1997b).
Following the recommendation (c) above, WSD nominated seven cut slopes, all
upstream and to the east of cut slope No. 6SE-D/CR135, for inclusion in the Landslip
Preventive Measures (LPM) Programme. In 1994 the slopes were selected by the
Interdepartmental Landslip Preventive Measures Committee for inclusion in the 1995/96
LPM Programme. A Stage 3 Study Report (GEO, 1997) was prepared by GEO in July 1997
and the LPM works including, inter alia, soil nailing, provision of surface drainage channels,
sprayed concrete surface cover, hydro-seeding and rock slope stabilization commenced in
August 1997 and were completed in January 1998. No works were undertaken on Slope
No.6SB~D/CR135.
In November 1997, Binnie Consultants Limited (BCL) prepared a draft working paper
for the Drainage Services Department (DSD), which summarised the findings of an
investigation into the impacts of overflow from WSD reservoirs and catchwaters on
downstream drainage systems, as part of the Tuen Mun and Sham Tseng Stormwater
- Ill Drainage Master Plan Study (BCL, 1997). In their report, BCL analysed the flows during
flooding incidents and assessed various options for reducing the impact of overflow together
with design improvements to the drainage system.
WSD intend to engage consultants in connection with a project to reconstruct
catchwaters. The scope of the work "will cover all slopes adjoining (sic) the catchwaters and
those potentially dangerous slopes downstream of the catchwaters, the failure of which will
affect Tuen Mun Road. Slope No. 6SE-D/CR135 will be included in the study" (WSD
1998a).
2.4
Previous Landslides
Inspection of aerial photographs dated 1975 shows a minor landslide above the
catch water cut slope about midway between overflow weirs C W and 'X 5 (Figure 3).
However, the failed material did not reach the catchwater. Minor erosion just below this area
is apparent on aerial photographs taken in 1978.
The Catchwater Study Second Report (GCO, 1982) describes that "during the 1981
rainy season part of the existing slide scar at CH6710-6720 was reactivated by a joint
controlled failure in rock behind the chunam. This blocked the catchwater and resulted in
considerable overflows along the two channels immediately upstream (CH6863 and CH6978).
Below the first overflow this caused some erosion to both cut and fill slopes adjoining the
Tuen Mun Highway. The increased discharges below the second overflow resulted in the
undermining of several squatter huts built over the gully below the Tuen Mun Highway. No
casualties were reported in these incidents". The location of this incident corresponds with
the location of a landslide (GCO Incident No. NT 23/81) recorded in GEO's Landslide
Incident Report database.
GEO's Landslide Incident Report database also contains a record of an incident (GEO
Incident No. MW93/6/56) which occurred on 16 June 1993 and involved blockage of the
trapezoidal drainage channel C D' (Figure 3). The incident was reportedly caused by "run off
washing down boulders in natural stream course". Unblocking of the channel and
reinstatement of the ground was recommended by GEO in the incident report.
GEO's Natural Terrain Landslide Inventory (NTLI) indicates there have been seven
natural terrain landslides on the hillsides of Ha Fa Shan above the Tai Lam Chung
Catchwater. The nearest failure to cut slope No. 6SE-D/CR135 was natural terrain landslide
C
D' which was first observed on aerial photographs taken in 1973 (Figure 3). The main scarp
was completely covered in vegetation and was less than 20 m wide. The debris trail was
about 50m long and was confined to a drainage line.
2.5
The geological memoir (Langford et al, 1989) and Sheet 6 of the Hong Kong
Geological Survey 1:20 000 scale Map Series (GCO, 1988) indicate that the site is underlain
by dacite, a very fine-grained equivalent of granodiorite.
- 112 -
There has been no previous ground investigation along the catchwater cut slope
No. 6SE-D/CR135. The subsurface conditions of the cut slope have been assessed from field
observations by HAP of the main scarps of the three landslides. All three main scarps
exposed weak to very weak, highly to completely decomposed dacite, and material of a
similar nature was also observed in the banks of the stream. In the bed of the stream,
moderately strong to strong, slightly decomposed dacite was exposed.
At the time of inspection by HAP on 3 July 1997, Slope Nos. 6SE-D/C64 and
6SE-D/F17 had been shotcreted as part of the urgent repair works. The closest borehole to the
site was drilled at the crest of Slope No. 6SE-D/C64 during the ground Investigation for Tuen
Mun Road Stage II works supervised by HyD consultants Scott Wilson Klrkpartick &
Partners (SWKP). The borehole encountered 3 m of colluvium described as a dense light
brown slightly gravelly sand with some cobbles overlying dense light brown and white clayey
gravelly sand (completely decomposed volcanics).
A geological cross-section from the Tai Lam Chung Catchwater to Tuen Mun Road is
presented in Figure 4.
3.
THE INCIDENT
3.1
The 1997 Landslides
The catchwater cut slope (No. 6SE-D/CR135) was affected by three landslides on
2 My 1997 (Plates 1 to 3). The locations of the landslides 'A', C B 9 and *C' are shown in
Figure 1. Landslide 'A' was about 1.5 m deep with an estimated volume of 280 m3 and
occurred on a part-vegetated, part-surfaced, 55°, 18m high soil cut slope. Landslides 'B' and
C
C' were about 1 m deep with estimated volumes of about 100 m3 and 35 m3. They occurred
on vegetated, 60°, soil cut slopes of around 10 m high and 5 m high respectively. The
material involved in each of the landslides was completely decomposed dacite. The travel
angle for all three landslides was in excess of 30° and the debris mobility is therefore not
unusual for rain-induced landslides in Hong Kong (Wong & Ho, 1996).
In May 1997, there was an incident (GEO Incident No. MW97/5/37) behind block M
of Allway Gardens, Tsuen Wan about 800 m upstream from overflow weir C X' on the Tai
Lam Chung Catchwater. GEO reported that ua minor landslip occurred on a slope above a
catchwater near Allway Gardens, resulting in blockage of the catchwater and overflowing at
the overflow weir (upstream of the landslip) down to a natural stream course. Debris and
minor landslips were found respectively inside and beside a large catchpit at the end of the
natural stream course" (GEO, 1998).
The previous reported incidents, together with the 1997 landslides, form a notable
cluster (Figure 3).
3.2
The Flooding Incident
A landslide warning was issued at 06:25 hours on 2 July 1997 and cancelled at
08:40 hours on 5 My 1997 and a flood warning was issued at 15:15 hours on 2 July 1997 and
cancelled at 09:30 hours on 3 Julv 1997.
- 113 -
The following sequence of events has been established based on reports produced by
HyD (1997a) and WSD (1997b). Three landslides occurred on cut slope No. 6SE-D/CR135
above the Tai Lam Chung Catchwater prior to midnight on 2 July 1997. The exact time of the
landslides and their sequence are not known. The landslide debris completely blocked the
catchwater. As a result of the blockage, water levels in the catchwater rose and water was
discharged through overflow weir *X' (Plate 4) into the stream. Flooding was first reported at
midnight on 2 July 1997 and continued until around 03:45 hours on 3 July 1997, at which
time, an emergency gang from WSD had cleared the landslide debris in the catchwater and reestablished flow.
The stormwater eroded the banks of the stream. Existing boulders in the bed of the
stream together with eroded soil, vegetation, loose masonry and rubbish from the adjacent
abandoned squatter platforms were all transported downstream (Plates 5, 6 & 7). As a result,
the whole length of the trapezoidal drainage channel C D' above the highway cut slope
(No. 6SE-D/C64) became completely blocked with debris (Plates 8 & 9). As a consequence,
there was an uncontrolled discharge of stormwater and debris down the highway cut slope on
either side of the cascade channel C E 5 (Plate 10), onto both carriageways of Tuen Mun Road.
The impact of debris striking the road and the scouring action of the water caused
damage to the surface of the east-bound carriageway of Tuen Mun Road for a length of
approximately 50 m (HyD, 1997a).
HyD contractors undertook urgent repair works to Tuen Mun Road between
02:15 hours and 15:15 hours on 3 July 1997. Apart from clearing the flood and washout
debris, they applied shotcrete to the damaged slope Nos. 6SE-D/C64 and 6SE-D/F17. The
west-bound carriageway was re-opened at 05:30 hours on 3 July 1997 and the east-bound
carriageway was re-opened at 15:30 hours on 3 July 1997. There was significant disruption to
traffic during the flooding incident and subsequent urgent repair works.
Following the incident, WSD contractors trimmed the main scarps of landslides C A 5 ,
C 9
B and C C', applied shotcrete protection and carried out repairs to drainage channels and the
concrete catchwater channel. HyD propose to undertake "survey, design and modification to
the inlet channel system" (HyD, 1997b). HyD also arranged for the stream to be cleared out
and the banks of the stream above the trapezoidal drainage channel to be protected with
shotcrete.
4.
RAINFALL
The nearest GEO automatic raingauge No. N03 is located at Tsuen Wan Treatment
Works, Shing Mun Road, about 3 km to the east of the flooding incident. The daily rainfall
recorded between 1 June and 11 July 1997 is shown in Figure 5a. There was 837 mm and
546 mm of rainfall in the 31 days and 15 days before the onset of the flooding incident
respectively. The hourly rainfall from 30 June to 3 July 1997 is shown in Figure 5b. There
was 364.5 mm and 182.5 mm of rainfall in the 24 hours and 12 hours before the flooding
respectively. Isohyets of rainfall prior to the flooding incident are shown in Figure 6. The
figure shows that the distribution of peak intensity rainfall was centred about 12 km northeast
of the site.
- 114 -
The estimated return periods for maximum rolling rainfall of selected durations based
on historical rainfall data at the Hong Kong Observatory (Lam & Leung, 1994) are presented
in Table 3. The maximum rolling 24-hour and 2-day rainfall totals were the most extreme
with corresponding return periods of about 9 years and 11 years respectively.
The 5-minute rainfall recorded at GEO raingauge No. N03 is shown in Figure 7. The
figure shows that the maximum intensity of the rainstorm between 17:00 hours on 2 July and
00:00 hours on 3 July was 8.5 mm and occurred between 21:35 hours and 21:40 hours on
2 July 1997.
The maximum daily rainfall in each month recorded at the GEO raingauge No. N03
since June 1983 is shown in Figure 8. Rainfall on 2 July 1997 was the highest daily total
recorded by the raingauge (HKO, 1997).
5.
CAUSES OF THE LANDSLIDES AND FLOODING INCIDENT
It is considered that the three landslides on the catchwater cut slope were probably
triggered by the heavy rainfall prior to midnight on 2 July 1997. The principal contributory
cause of the failures was infiltration of surface water into an oversteep slope.
The flooding incident probably developed in the following stages:
(a) Heavy rainfall in the 24 hours prior to the flooding incident
will have resulted in wetting of the hillside above the
catchwater and overland flows to the catchwater.
(b) Infiltration into both the natural terrain and cut slope above
the catchwater caused three landslides which completely
blocked the catchwater. The landslides were not caused by
an increase in water level in the catchwater. The close
proximity of the three landslides and their possible
occurrence within a short period of time would have made it
very difficult for the catchwater to self-clean through the
action of stormwater flow.
(c) Complete blockage of the catchwater by the landslide debris
resulted in the discharge of stormwater through the
overflow weir C X', causing the erosion of a large quantity of
debris from the bed and banks of the stream.
(d) The constriction made by the footbridge at the outfall of the
trapezoidal drainage channel C D', which discharges to
cascade channel C E' on the highway cut slope, was liable to
have contributed to the blockage of the trapezoidal channel
by flood debris.
(e) Blockage of the trapezoidal channel with debris prevented
drainage of stormwater and resulted in uncontrolled
- 115 -
discharge of stormwater and debris down the highway cut
slope onto both carriageways of Tuen Mun Road.
There is a notable cluster of past incidents near the location of the three landslides
which occurred on the catchwater cut slope in July 1997. Up to seven natural terrain
landslides have been identified on the hillside above the catchwater. Protective surfacing has
been applied to sections of the catchwater cut slope (No. 6SE-D/CR135) on at least four
occasions, one of which was at the location of GCO Incident No. NT23/81. There was also a
previous blockage of the trapezoidal drainage channel T)5 above highway cut slope
No. 6SE-D/C64 in June 1993 (GEO Incident No. MW93/6/56).
An issue of relevance in respect of the flooding incident is whether the stormwater
discharge at overflow weir 'X' was within, or in excess of, the design capacity of the weir.
WSD design information indicates that the design capacity of overflow weir 'X' was
sufficient to cope with a 1 in 200 year flood flow, provided that the catchwater was not
blocked. In the report on the incident by WSD (1997b), it is stated that "there are control
intakes and overflow weirs along the catchwater at chosen locations for discharging water in
excess of the designed capacity of the catchwater to the natural stream courses below to avoid
overtopping of the catchwater. The weirs will also discharge water to the natural stream
courses automatically when there is a sudden rise in the catchwater caused by blockages
resulting from the failure of the uphill slopes." The implication therefore, is that blockage of
the catchwater is taken into account in the hydraulic design.
It has been established that landslide debris completely blocked the catchwater
immediately downstream of overflow weir C X'. At 01:45 hours on 3 July 1997, HyD
engineers inspected the catchwater and reported that "there was overflowing at the overflow
weir and overtopping of the catchwater by about 50 mm over the adjacent road surface"
(HyD, 1997). Engineers from WSD inspected the overflow weir fiX' at 03:15 hours on
3 July 1997 and reported that "the water was observed to discharge freely from the overflow
weir and the maximum discharge capacity of the overflow weir had not been reached" (WSD,
1997b). Subsequent checks were made on the design capacity of the catchwater and overflow
weirs by WSD who reported that "based on our observation on the environmental evidence on
site and our detailed calculation for the run off from the catchment concerned at the time of
the incident which indicated that neither the capacity of the catchwater system nor that of the
overflow weirs had been exceeded, we still hold the view that there has been no uncontrolled
overtopping of the catchwater" (WSD, 1998b).
The results of a level survey of the catchwater carried out by HAP after the incident,
indicated that the height of the catchwater parapet wall above overflow weir 'X' was probably
sufficient to prevent overtopping of the catchwater in the event of blockage at the locations of
the three landslides (Figure 9). It is considered likely that overflow weirs *Y' and CZ* would
also have discharged stormwater as a result of the blockage of the catchwater. There was no
evidence of overtopping of the parapet wall of the catchwater upstream of overflow weir C X'.
In addition, the level survey determined that an 80 m long section of the catchwater
parapet wall immediately upstream from overflow weir *W is vulnerable to overtopping in
the event of a complete blockage of the catchwater along this section with landslide debris
(Figures 1 and 9).
- 116 -
6.
CONCLUSIONS
The flooding of Tuen Mun Road in the early hours of 3 July 1997 was caused by
uncontrolled discharge of stormwater and debris from a natural stream course above the road.
Stormwater was discharged into the stream via an overflow weir on the Tai Lam Chung
Catchwater which was blocked with debris from three landslides on the uphill cut slope
No. 6SE-D/CR135. The cut slope was not designed to or proved to comply with current
geotechnical safety standards. The stormwater eroded boulders and debris from the stream
bed and banks, which resulted in the blockage of a trapezoidal drainage channel at the crest of
the cut slope above the east-bound carriageway of Tuen Mun Road. The stormwater cascaded
down the cut slope and deposited water and debris onto the carriageways below. The access
footbridge over the outfall of the trapezoidal drainage channel may have contributed to the
blockage by constricting the flow of debris-laden stormwater.
7.
REFERENCES
Binnie Consultants Ltd. (1997). Agreement No. CE 63/96. Tuen Mun and Sham Tseng
Drainage Master Plan Study Working Paper no. 7 Draft Reservoir and Catchwater
Overflows. Binnie Consultants Ltd., 28 p., 2 figures and 4 appendices. (Unpublished).
Fugro Mouchel Rendel (1995).
Roadside Slope Inventory and Inspections Slope
No. 6SE-D/C64. Slope Record. 7 p. and Engineer Inspection Record, l i p .
Geotechnical Control Office (1982). Catchwater Study Second Report. Geotechnical Control
Office, Hong Kong, GCO Report 26/82. vol. 1, 20 p., 4 tables, 5 figures and
2 appendices, vol. 2, 6 appendices.
Geotechnical Control Office (1988). Yuen Long : solid and superficial geology. Hong Kong
Geological Survey. Map Series HGM 20. Sheet 6, 1:20 000 scale. Geotechnical
Control Office, Hong Kong.
Geotechnical Control Office (1989). Slope No. 6SE-D/C64, Stage 1 Study Report.
Geotechnical Control Office, Hong Kong, 4 p.
Geotechnical Engineering Office (1995). Guide to Slope Maintenance (Geoguide 5).
Geotechnical Engineering Office, Hong Kong, 92 p.
Geotechnical Engineering Office (1997). Slopes 6SE-D/C70, C128, C130 to C134, Tai Lam
Chung Catchwater, Section O, Chainage 7100 - 7950 Tsuen Wan. Stage 3 Study
Report. Geotechnical Engineering Office, Hong Kong, 9 p.
Geotechnical Engineering Office (1998).
1997 Landslip Investigation Consultancy.
Correspondence to HAP reference GCMW 2/A2/3 dated 24 February 1998.
Highways Department (1997). Report on the Damage to the Kowloon-Bound Carriageway of
the Tuen Mun Highway at Yau Kom Tau. Tsuen Wan on 2 - 3 July 1997. Highways
Department, New Territories Region, 19 p.
- 117 -
Hong Kong Observatory (1997). Monthly Summary of Daily Rainfall at GEO and Hong
Kong Observatory RaingauêsJifflBgJCong_Observatory.
selected locations in Hong Kong. Royal Observatory Hong Kong Technical Note.
no. 86, Hong Kong, 89 p.
Langford, R.L., Lai, K.W, Arthurton, R.S. & Shaw, R. (1989). Geology of the Western New
Territories. Hong Kong Geological Survey Memoir, no. 3, Geotechnical Control
Office, Hong Kong, 140 p.
Water Supplies Department & Geotechnical Control Office (1980). Catchwater Study First
Report, Hong Kong Government, 10 p., plus 2 appendices (Unpublished).
Water Supplies Department (1986). Flood Study on Tai Lam Chung Catchwater. Resource
Planning Report, no. RP4/84 (HS42), Hong Kong Government, 47 p.
Water Supplies Department (1997a). Existing utilities. Correspondence to HAP reference (3)
in WSD 674/16/2/5/90 Pt2 TJ(8) dated 21 August 1997.
Water Supplies Department (1997b). Report on Overflow of Water From the Tai Lam Chung
Catchwater at Chainage 6863 of Section LMNO on 3.7.1997. Technical Report.
Water Supplies Department, Hong Kong Government, 14 p.
Water Supplies Department (1998a). Closure of Tuen Mun Road. Correspondence to HAP
reference (75) in WSD 674/16/1/2/90 Pt. 2 dated 28 February 1998.
Water Supplies Department (1998b). Draft Landslip Study Report. Correspondence to Chief
Geotechnical Engineer Landslip Investigation. Geotechnical Engineering Office
reference (24) in WSD/MNW 1728/26/60 Pt. 3 dated 15 April 1998.
Wong H.N. & Ho, K.K.S. (1996). Travel distance of landslide debris. Proceedings of the
Seventh International Symposium on Landslides. Trondheim, Norway, vol. 1,
pp 417^ 422.
- 118 -
LIST OF TABLES
Table
No.
Page
No.
1
Summary of Site Development from Aerial Photograph
Interpretation
119
2
Summary of Information Sources
122
3
Maximum Rolling Rainfall at GEO Raingauge No. N03
for Selected Durations Preceding the 2 and 3 July 1997
Flooding Incident and The Corresponding Estimated
Return Periods
124
- 119 -
Table 1 - Summary of Site Development from Aerial Photograph Interrelation
(Sheet 1 of 3)
Year
Photographic
Reference No.
1954
Y02727/8
1963
Y08918/9
1974
8256/7
1975
11324/5
Altitude
Observations
These photographs were taken prior to
construction of the catchwater and show
that the natural hillside on the southern
flanks of Ha Fa Shan had several
prominent drainage lines.
The Tai Lam Chung Catchwater had
been constructed with a cut slope
directly above and a heavily vegetated
natural hillside beyond the cut slope.
Overflow weirs c W C X 9 and C Y' and C Z'
had been constructed on the catchwater
above the drainage lines. Bouldery
colluvium was evident in the drainage
line below overflow weir C X'. There
were a small number of squatter huts on
the hillsides below the catchwater to the
east and west of the natural stream
course.
1500ft
The catchwater cut slope remained
unchanged and the natural hillsides
above were heavily vegetated. The
village of Hon Man Tsuen had
developed by this time to the east and
west of the stream.
A U-channel drainage system was visible
above the catchwater cut slope. The
natural hillsides above the catchwater
were heavily vegetated and there was
evidence of a minor landslide above the
catchwater cut slope. In addition, a
section of the catchwater cut slope
downstream from overflow weir C X' had
a protective surfacing. The highway cut
slope was under construction and the
trapezoidal drainage channel C D' had not
been constructed at this time. The
relatively flat area immediately above
the highway cut slope had the form of a
colluvial debris fan.
- 120 -
Table 1 - Summary of Site Development from Aerial Photograph Interpretation
(Sheet 2 of 3)
Observations
Year
Photographic
Reference No.
Altitude
1977
17976
4000ft
1978
24035/6
4000ft
The footbridge over the east- and westbound carriageways to Nam Hoi Pui had
been constructed. Traffic was flowing
east and west on what is now the eastbound carriageway i.e. Tuen Mun Road
Stage I works had been completed. The
access footbridge over the outfall of the
trapezoidal drainage channel C D' had
been constructed by this time. There
was a small area of surface erosion on
the hillside below the minor landslide
which occurred in 1975. The U-channel
drainage system above the crest of the
catchwater cut slope was barely visible
through the dense vegetation.
1984
57517/8
4000ft
The construction of Tuen Mun Road
Stage II works had been completed and
east- and west-bound carriageways were
open to traffic.
1988
A3 174/5
10000ft
No changes were apparent.
1989
A16446/7
2500ft
1990
A20981/2
4000ft
1992
A3 1217/8
4000ft
The photographs show that there may
have been an uncontrolled discharge of
surface water down the highway cut
slope which resulted in a dark staining
on the surface of the slope.
There were no apparent changes to the
catchwater cut slope. The highway cut
slope had been constructed together with
the slope drainage system.
The
trapezoidal drainage channel C D' and the
cascade channel C E' were under
construction. The fill slope between the
east- and west-bound carriageways had
been constructed In the natural drainage
line and the culvert C F' under the Tuen
Mun Road was in place.
- 121 -
Table 1 - Summary of Site Development from Aerial Photograph Interpretation
(Sheet 3 of 3)
Year
Photographic
Reference No.
Altitude
Observations
1993
A35991/2
4000ft
The dark staining on the surface of the
highway cut slope was not evident.
1994
A38 170/1
5000ft
1996
CN15761/2
4000ft
- 122 -
Table 2 - Summary of Information Sources (Sheet 1 of 2)
Source
Information Obtained
Geotechnical Information Unit Stage 3 Study Report. Slopes 6SE-D/C70, C128,
(GIU) at the Civil Engineering C130 to C134 Tai Lam Chung Catchwater Section 0
Chainage 7100 to 7950, Tsuen Wan, prepared by
Department (CED) Library.
GEO Design Division.
Slope Files and District Files from (a) Stage 1 Study for Slope Nos. 6SE-D/C135 and
6SE-D/C64.
Mainland West Division and Files
from Design Division of GEO.
(b) GEO Incident Report No. MW93/6/56.
(c) Provisional Brief for Reconstruction of I
Catchwaters in Tai Lain Chung, Hong Kong
Island and Lantau Island.
(d) Tuen Mun and Sham Tseng Drainage Master Plan
Study.
GEO Publications, Reports, Maps (a) Yuen Long : Solid and superficial geology, Hong
Kong Geological Survey, Map Series HGM 20,
and Memoirs.
Sheet 6, 1:20 000 scale.
(b) Geology of the Western New Territories, Hong
Kong Geological Survey Memoir No. 3.
(c) Catchwater Study First Report.
(d) Catchwater Study Second Report, GEO Report
No. 26/82, vols 1 & 2.
GEO Planning Division.
GEO Slope Safety Division.
Aerial Photographs from 1954, 1963, 1974, 1975,
1976, 1977, 1978, 1984, 1988, 1989, 1990, 1992,
1993,1994 and 1996.
(a) SIRST Report on Slope No.
6SE-D/CR135.
(b) Slope Record and Engineers Inspection Record
for slope No. 6SE-D/C64 prepared by Fugro
Mouchel Rendel for HyD.
GEO Landslide Incident Report Details of past landslides reported to GEO.
Database.
Lands Department and District (a) Land Status.
Lands Office (DLO).
(b) Maintenance Responsibility.
- 123 -
Table 2 - Summary of Information Sources (Sheet 2 of 2)
Source
Water
(WSD).
Information Obtained
Supplies
Department (a) Summary Report on 3 July 1997 Landslip
Incident on Tai Lam Chung Catchwater Section
NChainage 6700 to 6863.
(b) Full Incident Report.
(c) Flood Study on Tai Lam Chung Catchwater.
Resources Planning Report No. RP4/84 (HS42).
Drainage
(DSD).
Services
Department Existing Utility Information.
LPM / SIRST / SIFT Databases.
Registration Status of Slope Nos. 6SB-D/CR135,
6SE-D/C64 and 6SE-D/F17.
Highways Department (HyD).
(a) Report on the Drainage to the Kowloon-Bound
Carriageway Surface of Tuen Mun Highway At
Yau Kom Tau, Tsuen Wan on 2 and 3 July 1997.
(b) Video of TVB Pearl News Report on the
Flooding Incident.
Hong Kong Observatory (HKO).
(a) Daily rainfall data between June 1983 and July
1997.
(b) Isohyets of rainfall between 00:00
2 July and 00:00 hours 3 July 1997.
hours
- 124 -
Table3 - Maximum Rolling Rainfall at GEO Raingauge No. N03 for Selected
Durations Preceding the 2 and 3 July 1997 Flooding Incident and The
Corresponding Estimated Return Periods
Maximum
Rolling
Rainfall (mm)
End of Period
Estimated Return
Period (Years)
5 minutes
13.5
2
15 minutes
35
6
1 hour
89
5
2 hours
120
4
4 hours
129.5
1
12 hours
197.5
3
24 hours
364.5
9
2 days
441
11
4 days
459.5
6
7 days
468
4
15 days
546
3
31 days
837
5
Duration
Notes: (1) Return periods were derived from the Gumbel equation and data published
in Table 3 of Lam & Leung (1994).
(2) Maximum rolling rainfall was calculated from 5 -minute data for durations
up to one hour and from hourly data for longer rainfall durations.
- 125 -
LISTOFFIGTTRRS
Figure
No
-
Page
No.
1
Site Location Plan
126
2
Site Plan and Incident Details
127
3
Site History and Land Status
128
4
Geological Cross-section
129
5
130
6
Isohyets of Rainfall between 00:00 Hours on 2 July and 00:00 Hours on
3 July 1997
131
7
Rainfall Recorded at GEO Raingauge No. N03 at 5-minute Intervals
on 3 July 1997
132
8
Maximum Daily Rainfall in Each Month at GEO Raingauge No. N03
133
9
Interpretation of Level Survey Results along the Tai Lam Chung Catchwater
134
- 126 -
o
o
o
4~>
c/5
-127-
w
3
Q
d
O
a
d
Kl
d
cd
sO
•tt
C/5
toD
- 128 -
CO
4.1
cti
.4-*
CO
cd
K->
J^
o
(75
- 129 -
s
S-i
CU
O
C
P
j^
-i
03
T3
O
JD
S
p
el.
J.
a
o
O
<D
00
CO
C/3
O
u
13
o
.*•*
bo
0)
M
bD
p
o
a
.2
o
<D
W
I
<W
CO
O
1-1
O
o
«g
S3
00
O
CJ
.2
t3
o
5
«~l
o
•o
bfl
0>
CD
O
- 130 837 mm of rainfall recorded in the 31 days before the reported time of the incident
546 mm of rainfall recorded in the 15 days before the reported time of the incident
Lg
H
_H
500
Reported Time of the Flooding Incident
450 400 350 300
£ 250 KJ
£» 200 Q
150 100 -
50 -
. I. I.. ..ll.. i.l ILJL JJ
0
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
3
1
5
7
9
11
July 1997
June 1997
(a) Daily Rainfall Recorded between 1 June and 12 July 1997
364 5 mm of rainfall recorded in the 24-hours before the reported time of the incident
182 5 mm of rainfall recorded in the 12-hours before the reported time of the incident
h
J\J -
H
Reported Time of the Flooding Incident
at 00 00 hours on 3 July 1997
45 40-
?
6
a
35
~
30
"
'
g 25 .
% 20"
o
K 15-
J
10-
t i H
Ilil Illli , 1 1
H
5
.
0-
000
1200
30 June 1997
1
000
ll
ll Illllll..
1200
1 July 1997
1
J
11
000
1200
1200
000
2 July 1997
H«
i
3 July 1997
(b) Hourly Rainfall Recorded between 30 June and 3 July 1997
000
131 -
2
H-j
m
o
co
O
ffi
o
o
CD
O
a
o
00
i-(
^
O
o
?.
o
o
a
o
<D
^
4->
0)
JD
C4H
o
CO
4^
0)
P^
^!
O
CO
.SP
£
132 -
CO
G
O
s
o
Z
o
2
CD
bD
3bD
O
w
o
CD
"S
8
0)
eg
I
(^
I
l>
I
OD
- 133 -
o
a
o
O
Tf
(WUl)
- 134CD
a
—n
N
^
v
\
P-
>
J5
*
*
fe
0
\
1)
bJO
M
.^
S- o
r
2
r-
CD
o
o
A-
^
^
b
R
I
aq
|S
x
i
I
"3
||
i
_
\
O
1
ect
u< fi
> 2
o o
'
;
CD
L
I
'
c*
strz" l
|Pi
*
w
^ f-I
CD
03
cQ
^ S *^
S
c2 c
*l
ft
o
~~ oo
I
^
\
QJ 1^ J.^1
r\
i_<
'
5^ (D
o
o
CO
"" OO
1 |
P"
1«
"s 'S "5
\
—
J5
i J
^ a,
-J
\
»
f>
0
»
1«J5!
l
ft
r>
CD
ci
wr.
»~l
XI
CD
"S
CL
o
o
13
Cfl
4_>
3
00
CD
CD
bo
<
r*
H
o
o
<°
w)4u ^ \
p-
W
\
III
^
"H
bD
C
3
X3
0
_ o
of catchw
n hn
u
S
S
a,
r~*
4->
OJ
U
vo
1
y
CD ^^
CQ
"~ ON
i
•gf V
o
~~ 0
o
O "S
o
o
\
*
o
S
o
"CD
S
* 1
11
03
O
"" 0
t
^
}_,
(U
c3
^
O
o
\
\
,
\
£S
«
13
oJ oJ
pO,
5 Ox
-2 8
S
\
°" V
> L
^
O
^
*.
^""
S vo
03 v^
2- ,j
^
i
|.a
o
g-|
"c3
2
°2?— \
bO
^ /"""N
«
m
| o
s
i
\\
"5 * o p.
° ^ -C
CD
cu
o
\
11
o-
i
O
a
•
^
2
^
\
\
^g
c>
§
c 2 "3L -f
S « P^
.2 js § *
\
»
CA
0 "3 ° a
\
w
11
IH
~i
.2 ^ ~*
0
t-4
^—JL\
•? 1 1 1
——. —. __
T3
CD
bO "CD
*••«».
^
»-. ^
*S S
> */^
^
£L
P ex ^
*sO
>
0
C
'^
i
<—'
a>
,.
"*
^"^ ••"<
O
c/2
n»t
«•*
^
,
o
«n
"" «n
^0
v
"^
^
\
^
k
'
1
t
Hi-
o
CD
j>
CD
^
xr
O
<*H
vO
^ j-j csi CD
ch
TK— CU,
hi)
CD
>
o
__ o
\o
O o (L)
O
OO 2^ rt **•• •'-"
-H t-i "*~" -5J "rj
5
6 g z: o/S °* 2
\
tf -
o
Oô
— O
d
2
2
2
2
(ddûo^H
^
^
CD
W)
CD
J
as
<u
*H
&
- 135 -
LIST OF PLATES
Plate
No.
Page
No-
1
View Showing Overflow Weir 'X' and Landslide 'A' on the Tai
Lam Chung Catchwater Cut Slope No. 6SE-D/CR135 (Photograph
136
2
Landslide 'B' on the Tai Lam Chung Catchwater Cut Slope No.
6SE-D/CR135 (Photograph Taken on 7 August 1997)
136
3
Landslide 'C' on the Tai Lam Chung Catchwater Cut Slope No.
6SE-D/CR135 (Photograph Taken on 7 August 1997)
137
4
View Down the Stream from Overflow Weir 'X' on the Tai Lam
Chung Catchwater (Photograph Taken on 10 July 1997)
137
5
View Down the Stream Showing the Flood Debris (Photograph
138
6
Erosion of Banks of the Stream (Photograph
10 July 1997)
Taken on
138
7
Flood Debris Accumulated at a Break in Slope of the Stream
(Photograph Taken on 7 August 1997)
139
8
Trapezoidal Drainage Channel 'D' Completely Blocked with Flood
Debris (Photograph Taken on 4 July 1997)
139
9
Cascade Channel 'E' down Cut Slope No. 6SE-D/C64 Prior to
Debris Clearance (Photograph Taken on 4 July 1997)
140
10
Trapezoidal Drainage Channel 'D' above Cut Slope
No. 6SE-D/C64 after Urgent Repair Works (Photograph Taken on
7 August 1997)
141
11
View of Fill Slope No. 6SE-D/F17 between the East- and WestBound Carriageways of Tuen Mun Road (Photograph Taken
on 4 July 1997)
142
- 136 -
Plate 1 - View Showing Overflow Weir 'X' and Landslide 'A1 on the
Tai Lam Chung Catchwater Cut Slope No. 6SE-D/CR135
Plate 2 Landslide 'B' on
the Tai Lam
Chung Catchwater
Cut Slope No.
6SE-D/CR135
(Photograph
Taken on
7 August 1997)
- 137 -
Plate 3 - Landslide 'C' on the Tai Lam Chung Catchwater Cut Slope
No. 6SE-D/CR135 (Photograph Taken on 7 August 1997)
Plate 4 - View Down the Stream from Overflow Weir 'X' on the Tai Lam
Chung Catchwater (Photograph Taken on 10 July 1997)
- 138 -
Plate 5 - View Down the Stream Showing the Flood Debris
Plate 6 - Erosion of Banks of the Stream (Photograph Taken on 10 July 1997)
- 139-
^
^
S£
O ON
3§
Q H
13 -|
J3
O
U o
.S .2
^ .^
Q ^
ZH
Q
O
fe
1
:
. -
,2 ON
& ON
G ^
•- to
'TO
too
•s §
II
Is
It
"8 PM
Q ^
11
PH GO
—
-
- 140 -
Plate 9 - Cascade Channel 'F down Cut Slope No. 6SE-D/C64
Prior to Debris Clearance (Photograph Taken on 4 July 1997)
- 141 -
Plate 10 - Trapezoidal Drainage Channel 'D' above Cut Slope
No. 6SE-D/C64 after Urgent Repair Works (Photograph
Taken on 7 August 1997)
- 142-
<D
00
T— (
:
E
«4H
0
1
>
/:
c
.
:
-o
|
:
^
32
—: --
Carriageways of
OH
J2 ----
'-
0\
o\
r—<
1
1
1
1
J
J
i:
_
>,
3
H->
^t
-
0
Geotechnlcal Manual for Slopes, 2nd Edition (1984), 295 p.
(English Version), (Reprinted, 1997).
HK$90
(US$20)
^fi( 1 998) > 308MC 1
HK$90
(US$20)
Guide to Retaining Wall Design, 2nd Edition (1993), 258 p.
(Reprinted, 1998).
Geoguide 1
HK$60
(US$13)
Guide to Site Investigation ( 1 987), 359 p. (Reprinted, 1996).
Geoguide 2
HK$83
(US$17.5)
Guide to Rock and Soil Descriptions (1988), 186 p. (Reprinted,
1997).
Geoguide 3
HK$58
(US$12.6)
Guide to Cavern Engineering (1992), 159 p. (Reprinted, 1994).
Geoguide 4
HK$36
(US$13.5)
Guide to Slope Maintenance (1995), 91 p. (English Version).
Geoguide 5
HK$30
(US$6.5)
HK$40
(US$7.5)
Layman's Guide to Slope Maintenance, 2nd Edition (1998),
54 p. (Bilingual).
Free
Model Specification for Prestressed Ground Anchors, 2nd
Edition (1989), 164 p. (Reprinted, 1997).
Geospec 1
HK$62
(US$11)
Model Specification for Reinforced Fill Structures (1989),
135 p. (Reprinted, 1997).
Geospec 2
HK$58
(US$10.5)
Mid-levels Study : Report on Geology, Hydrology and Soil
Properties (1982), 265 p. plus 54 drgs. (Reprinted, 1997).
-
Prediction of Soil Suction for Slopes in Hong Kong, by M.G.
Anderson (1984), 242 p. (Reprinted, 1996).
GCO Publication
No. 1/84
(Superseded by GCO Publication No. 1/85)
GCO Publication
No. 2/84
HK$534
(US$86)
HK$132
(US$24)
(Superseded by Geospec 1 )
GCO Publication
No. 3/84
Review of Superficial Deposits and Weathering in Hong Kong,
by J.D. Bennett (1984), 58 p. (Reprinted, 1993).
GCO Publication
No. 4/84
HK$40
(US$8)
Review of Hong Kong Stratigraphy, by J.D. Bennett (1984),
86 p.
GCO Publication
No. 5/84
HK$25
(US$5.5)
Review of Tectonic History, Structure and Metamorphism of
Hong Kong, by J.D. Bennett (1984), 63 p.
GCO Publication
No. 6/84
HK$20
(US$5)
(Superseded by GCO Publication No. 1/88)
GCO Publication
No. 1/85
Groundwater Lowering by Horizontal Drains, by D.J. Craig &
I. Gray ( 1 985), 123 p. (Reprinted, 1 990).
GCO Publication
No. 2/85
(Superseded by GEO Report No. 9)
GCO Publication
No. 1/88
Review of Design Methods for Excavations (1990), 187 p.
(Reprinted, 1996).
GCO Publication
No. 1/90
HK$40
(US$10)
Foundation Properties of Marble and Other Rocks in the Yuen
Long - Tuen Mun Area (1990), 1 17 p.
GCO Publication
No. 2/90
HK$58
(US$10)
HK$74
(US$ 1 2)
Review of Earthquake Data for the Hong Kong Region ( 1 99 1 ), GCO Publication HK$42
115 p.
No. 1/91
(US$11.5)
Review of Granular and Geotextile Filters ( 1 993), 1 4 1 p.
GEO Publication HK$32
No. 1/93
(US$19)
Pile Design and Construction (1996), 348 p. (Reprinted, 1997).
GEO Publication
No. 1/96
HK$62
(US$13.5)
Report on the Kwun Lung Lau Landslide of 23 July 1994, 2
Volumes, 400 p. (English Version), (Reprinted, 1996).
-
Free
Report on the Fei Tsui Road Landslide of 13 August 1995,
2 Volumes, 81 p. (Bilingual).
-
Free
^5f|
HB
'
Report on the Shum Wan Road Landslide of 13 August 1995,
2 Volumes, 63 p. (Bilingual).
Free
What to Do When You Receive a Dangerous Hillside Order
(1996), 16 p. (Bilingual).
Free
B
>
[(1996)
8=0
(Hong Kong) Rainfall and Landslides in 1984, by J. Premchitt
(1991), 91 p. plus 1 drg. (Reprinted, 1995).
GEO Report
No. 1
HK$118
(US$17.5)
GEO Report
No. 2
HK$126
(US$20)
GEO Report
No. 3
HK$126
(US$20)
Hong Kong Rainfall and Landslides in 1987, by J. Premchitt
GEO Report
No. 4
HK$122
(US$19.5)
Hong Kong Rainfall and Landslides in 1988, by J. Premchitt
GEO Report
No. 5
HK$106
(US$ 16)
Hong Kong Rainfall and Landslides in 1989, by K.L. Siu
GEO Report
No. 6
HK$126
(US$20)
Aggregate Properties of Some Hong Kong Rocks, by T.Y.
Man, A. Cipullo, A.D. Burnett & J.M. Nash (1992), 212 p.
(Reprinted, 1995).
GEO Report
No. 7
HK$120
(US$19.5)
Foundation Design of Caissons on Granitic and Volcanic
Rocks, by T.Y. Man & G.E.Powell (1991), 85 p. (Reprinted,
1995).
GEO Report
No. 8
HK$62
(US$10.5)
Bibliography on the Geology and Geotechnical Engineering of
Hong Kong to December 1991, by E.W. Brand (1992), 186 p.
(Superseded by GEO Report No.39)
GEO Report
No. 9
Bibliography on Settlements Caused by Tunnelling, by E.W.
Brand (1992), 50 p. (Reprinted, 1995). (Superseded by GEO
Report No.51)
GEO Report
No. 10
HK$48
(US$8.5)
Direct Shear Testing of a Hong Kong Soil under Various
Applied Matric Suctions, by J.K. Gan & D.G. Fredlund (1992),
GEO Report
No. 11
HK$136
(US$21.5)
Rainstorm Runoff on Slopes, by J. Premchitt, T.S.K. Lam, J.M.
Shen and H.F. Lam (1992), 211 p. (Reprinted, 1995).
GEO Report
No. 12
HK$ 121
(US$19.5)
Mineralogical Assessment of Creep-type Instability at Two
Landslip Sites, by T.Y. Man (1992), 136 p. (Reprinted, 1995).
GEO Report
No. 13
HK$87
(US$15)
Hong Kong Rainfall and Landslides in 1990, by K.Y. Tang
(1992),78p.plusldrg. (Reprinted, 1995).
GEO Report
No. 14
HK$112
(US$17)
Assessment of Stability of Slopes Subjected to Blasting GEO Report
Vibration, by H.N. Wong & P.L.R. Pang (1992), 112 p. No. 15
(Reprinted, 1995).
HK$75
(US$12)
Earthquake Resistance of Buildings and Marine Reclamation
Fills in Hong Kong, by W.K. Pun (1992), 48 p. (Reprinted,
1995).
GEO Report
No. 16
HK$48
(US$8.5)
Review of Dredging Practice in the Netherlands, by S.T.
Gilbert & P.W.T. To (1992), 112 p. (Reprinted, 1995).
GEO Report
No. 17
HK$76
(US$12)
Backfilled Mud Anchor Trials Feasibility Study, by C.K. Wong
& C.B.B. Thorley (1992), 55 p. (Reprinted, 1995).
GEO Report
No. 18
HK$50
(US$9)
A Review of the Phenomenon of Stress Rupture in HDPE
Geogrids, by G.D. Small & J.H. Greenwood (1993), 68 p.
(Reprinted, 1995).
GEO Report
No. 19
HK$56
(US$9.5)
Hong Kong Rainfall and Landslides in 1991, by N.C. Evans
GEO Report
No. 20
HK$111
(US$ 16.5)
Horizontal Subgrade Reaction for Cantilevered Retaining Wall
Analysis, by W.K. Pun & P.L.R. Pang (1993), 41 p. (Reprinted,
1998).
GEO Report
No. 21
HK$50
(US$9.5)
Report on the Rainstorm of 8 May 1992, by N.C. Evans (1993),
109 p. plus 2 drgs. (Reprinted, 1995).
GEO Report
No. 22
HK$ 126
(US$20)
Effect of the Coarse Fractions on the Shear Strength of
Colluvium, by T.Y. Man & K.Y. Tang (1993), 223 p.
(Reprinted, 1995).
GEO Report
No. 23
HK$126
(US$20)
The Use of PFA in Reclamation, by J. Premchitt & N.C. Evans
(1993), 59 p. (Reprinted, 1995).
GEO Report
No. 24
HK$52
(US$9)
Report on the Rainstorm of May 1982, by M.C. Tang (1993),
129 p. plus 1 drg. (Reprinted, 1995).
GEO Report
No. 25
HK$135
(US$21)
Report on the Rainstorm of August 1982, by R.R. Hudson
GEO Report
No. 26
HK$118
(US$17.5)
Landslips Caused by the June 1983 Rainstorm, by E.B. Choot
(1993), 124 p. (Reprinted, 1995).
GEO Report
No. 27
HK$83
(US$ 13)
Factors Affecting Sinkhole Formation, by Y.C. Chan (1994),
GEO Report
No. 28
HK$40
(US$7.5)
Classification and Zoning of Marble Sites, by Y.C. Chan
(1994), 37 p. (Reprinted, 1995).
GEO Report
No. 29
HK$40
(US$7.5)
Hong Kong Seawall Design Study, by P.M. Aas & A. Engen
(1993), 94 p. (Reprinted, 1995).
GEO Report
No. 30
HK$68
(US$ 11)
Study of Old Masonry Retaining Walls in Hong Kong, by Y.C.
Chan (1996), 225 p.
GEO Report
No. 31
HK$130
(US$21)
Karst Morphology for Foundation Design, by Y.C. Chan &
W.K. Pun (1994), 90 p. plus 1 drg. (Reprinted, 1995).
GEO Report
No. 32
HK$118
(US$17.5)
An Evaluation of the Suitability of Decomposed Granite as
Foundation Backfill for Gravity Seawalls in Hong Kong, by
E.B. Choot (1993), 34 p. (Reprinted, 1995).
GEO Report
No. 33
HK$38
(US$7)
A Partial Factor Method for Reinforced Fill Slope Design, by
H.N. Wong (1993), 55 p. (Reprinted, 1995).
GEO Report
No. 34
HK$50
(US$9)
Hong Kong Rainfall and Landslides in 1992, by P.K.H. Chen
(1993), 201 p. plus 2 drgs. (Reprinted, 1995).
GEO Report
No. 35
HK$167
(US$25.5)
Methods of Test for Soils in Hong Kong for Civil Engineering
Purposes (Phase I Tests), by P.Y.M. Chen, 1996 Edition, 90 p.
GEO Report
No. 36
HK$22
(US$5.5)
Creep, Stress Rupture and Hydrolysis of Polyester Reinforced
Geogrids, by J.H. Greenwood (1995), 67 p.
GEO Report
No. 37
HK$38
(US$7.5)
Skin Friction on Piles at the New Public Works Central
Laboratory, by J.Premchitt, I. Gray &K.K.S. Ho (1994), 158 p.
(Reprinted, 1995).
GEO Report
No. 38
HK$97
(US$16.5)
Hong Kong to May 1994, by E.W. Brand (1994), 202 p.
(Reprinted, 1995). (Superseded by GEO Report No.50)
GEO Report
No. 39
HK$ 118
(US$19)
Hydraulic Fill Performance in Hong Kong, by C.K. Shen &
K.M. Lee (1995), 199 p.
GEO Report
No. 40
HK$90
(US$16)
Mineralogy and Fabric Characterization and Classification of
Weathered Granitic Rocks in Hong Kong, by T.Y. Lfan (1996),
158 p.
GEO Report
No. 41
HK$70
(US$ 13.5)
Performance of Horizontal Drains in Hong Kong, by R.P.
Martin, K.L. Siu & J. Premchitt (1995), 109 p.
GEO Report
No. 42
HK$65
(US$17.2)
Hong Kong Rainfall and Landslides in 1993, by W.L. Chan
(1995), 214 p. plus 1 drg.
GEO Report
No. 43
HK$110
(US$18.5)
General Report on Landslips on 5 November 1993 at
Man-made Features in Lantau, by H.N. Wong & K.K.S. Ho
(1995), 78 p. plus 1 drg.
GEO Report
No. 44
HK$64
(US$17)
Gravity Retaining Walls Subject to Seismic Loading, by Y.S.
Au-Yeung&K.K.S.Ho(1995),63p.
GEO Report
No. 45
HK$40
(US$8)
Direct Shear and Triaxial Testing of a Hong Kong Soil under
Saturated and Unsaturated Conditions, by J.K.M. Gan & D.G.
Fredlund (1996), 217p.
GEO Report
No. 46
HK$65
(US$ 12.5)
Stability of Submarine Slopes, by N.C. Evans (1995), 51 p.
GEO Report
No. 47
HK$46
(US$8.5)
Strength Development of High PFA Content Concrete, by
W.C.Leung&W.L.Tse(1995),84p.
GEO Report
No. 48
HK$60
(US$10.5)
AAR Potential of Volcanic Rocks from Anderson Road
Quarries, by W.C. Leung, W.L. Tse, C.S. Mok & S.T. Gilbert
(1995), 78 p.
GEO Report
No. 49
HK$58
(US$ 10)
Hong Kong to March 1996, by E.W. Brand (1996), 111 p.
GEO Report
No. 50
HK$45
(US$9)
Bibliography on Settlements Caused by Tunnelling to March
1996, by E.W. Brand (1996), 70 p.
GEO Report
No. 51
HK$31
(US$6.5)
Investigation of Some Major Slope Failures between 1992 and
1995, by Y.C. Chan, W.K. Pun, H.N. Wong, A.C.O. Li &K.C.
Yeo (1996), 97 p.
GEO Report
No. 52
HK$44
(US$8.5)
Environmental Aspects of Using Fresh PFA as Fill in
Reclamation, by K.S. Ho & P.Y.M. Chen (1996), 46 p.
GEO Report
No. 53
HK$30
(US$5.5)
Hong Kong Rainfall and Landslides in 1994, by W.L. Chan
(1996), 161 p. plus 1 drg.
GEO Report
No. 54
HK$70
(US$13.5)
Conventional and CRS Rowe Cell Consolidation Test on Some
Hong Kong Clays, by J. Premchitt, K.S. Ho & N.C. Evans
(1996), 93 p.
GEO Report
No. 55
HK$35
(US$7)
Application of Prescriptive Measures to Soil Cut Slopes, by
H.N. Wong & L.S. Pang (1996), 52 p.
GEO Report
No. 56
HK$12
(US$3.5)
Study of Rainfall Induced Landslides on Natural Slopes in the
Vicinity of Tung Chung New Town, Lantau Island, by
C.A.M. Franks (1998), 102 p. plus 3 drgs.
GEO Report
No. 57
HK$264
(US$39.5)
Tsing Shan Debris Flow and Debris Flood, by J.P. King
(1998), 215 p. plus 9 drgs. under preparation.
GEO Report
No. 58
HK$760
(US$107)
Hong Kong Rainfall and Landslides in 1995, by C.K.L. Wong
(1997), 125 p. plus Idrg.
GEO Report
No. 59
HK$70
(US$14.5)
Assessment of Geological Features Related to Recent
Landslides in Volcanic Rocks of Hong Kong Phase 2A - Chai
Wan Study Area, by S.D.G. Campbell & N.P. Koor (1998),
78 p. plus 6 drgs.
GEO Report
No. 60
HK$296
(US$43.5)
Factual Report on the November 1993 Natural Terrain
Landslides in Three Study Areas on Lantau Island, by H.N.
Wong, Y.M. Chen & K.C. Lam (1997), 42 p.
GEO Report
No. 61
HK$92
(US$13.5)
Areal Extent of Intense Rainfall in Hong Kong 1979 to 1995,
byA.W. Malone (1997), 85 p.
GEO Report
No. 62
HK$43
(US$8.5)
A Review of Some Drained Reclamation Works in Hong
Kong, by J.S.M. Kwong (1997), 53 p.
GEO Report
No. 63
HK$36
(US$6.5)
A Study of Hydraulic Fill Performance in Hong Kong Phase 2, by C.K. Shen, K.M. Lee & X.S. Li (1997), 265 p.
GEO Report
No. 64
HK$150
(US$25)
Seismic Hazard Analysis of the Hong Kong Region, by C.F.
Lee, Y.Z. Ding, R.H. Huang, Y.B. Yu, G.A. Guo, P.L. Chen
& X.H. Huang (1998), 145 p. (Bilingual).
GEO Report
No. 65
HK$80
(US$15.5)
i^M^f/f' ^'W^- - TM*
- iUtiIW«fi«1i(1998)
Mineralogical and Fabric Characterization and Classification of GEO Report
Weathered Volcanic Rocks in Hong Kong, by T.Y. Man (1998), No. 66
113 p.
HK$106
(US$17)
Assessment of Geological Features Related to Recent
Landslides in Volcanic Rocks of Hong Kong Phase 2BAberdeen Study Area, by C.A.M. Franks, S.D.G. Campbell &
W.W.L. Shum (1998), under preparation.
GEO Report
No. 67
The New Priority Classification Systems for Slopes and
Retaining Walls, by C.K.L. Wong (1998), 117 p.
GEO Report
No. 68
HK$66
(US$11.5)
Diagnostic Report on the November 1993 Natural Terrain
Landslides on Lantau Island, by H.N. Wong, K.C. Lam &
K.K.S. Ho (1998), 98 p. plus 1 drg.
GEO Report
No. 69
HK$90
(US$17)
Hong Kong Rainfall and Landslides in 1996, by C.K.L. Wong
(1998), 84 p. plus 1 drg.
GEO Report
No. 70
HK$ 112
(US$20)
Site Characterisation Study - Phases 1 and 2, by N.P. Koor
(1998), under preparation.
GEO Report
No. 71
Long-term Consolidation Tests on Clays from the Chek Lap
Kok Formation, by D.O.K. Lo & J. Premchitt (1998), 89 p.
GEO Report
No. 72
The Natural Terrain Landslide Study Phases I and n, by N.C.
Evans, S.W. Huang & J.P. King (1998), under preparation.
GEO Report
No. 73
Natural Terrain Landslide Study the Natural Terrain Landslide
Inventory, by J.P. King (1998), under preparation.
GEO Report
No. 74
Landslides and Boulder Falls from Natural Terrain : Interim
Risk Guidelines, by ERM-Hong Kong, Ltd (1998), 183 p.
GEO Report
No. 75
Report on the Landslides at Hut No. 26 Kau Wa Keng Upper
Village of 4 June 1997, by Halcrow Asia Partnership Ltd.
(1998), 100 p. (Bilingual), under preparation.
GEO Report
No. 76
N H 0 B 7W?S±*!f 26MUjMftI WM
Report on the Landslide at Ten Thousand Buddhas' Monastery GEO Report
of 2 July 1997, by Halcrow Asia Partnership Ltd. (1998), 96 p. No. 77
(Bilingual), under preparation.
ftfi^' &
HK$34
(US$7.5)
HK$94
(US$17.5)
Report on the Ching Cheung Road Landslide of 3 August
1997, by Halcrow Asia Partnership Ltd. (1998), 142 p.
(Bilingual).
GEO Report
No. 78
HK$256
(US$38.5)
Investigation of Some Selected Landslide Incidents in 1997
(Volume 1), by Halcrow Asia Partnership Ltd. (1998), 142 p.
GEO Report
No. 79
HK$192
(US$30)
Feasibility Study for QRA of Boulder Fall Hazard in Hong
Kong, by ERM-Hong Kong, Ltd (1998), 61 p.
GEO Report
No. 80
HK$58
(US$10.5)
Geotechnical Area Studies Programme - Hong Kong and
Kowloon (1987), 170 p. plus 4 maps.
GASP I
HK$240
(US$40)
Geotechnical Area Studies Programme - Central New
Territories (1987), 165 p. plus 4 maps.
GASPn
HK$150
(US$25)
Geotechnical Area Studies Programme - West New Territories
( 1 987), 1 55 p. plus 4 maps.
GASP HI
HK$ 1 50
(US$25)
Geotechnical Area Studies Programme - North West New
GASP IV
HK$150
(US$25)
Geotechnical Area Studies Programme - North New Territories
( 1 988), 1 34 p. plus 3 maps.
GASP V
HK$ 1 50
(US$25)
Geotechnical Area Studies Programme - North Lantau (1988),
124 p. plus 3 maps.
GASP VI
HK$150
(US$25)
Geotechnical Area Studies Programme - Clear Water Bay
(1988), 144 p. plus 4 maps.
GASPVH
HK$150
(US$25)
Geotechnical Area Studies Programme - North East New
GASPVIH
HK$150
(US$25)
Geotechnical Area Studies Programme - East New Territories
(1988), 141 p. plus 4 maps.
GASP K
HK$150
(US$25)
Geotechnical Area Studies Programme - Islands (1988), 142 p.
plus 4 maps.
GASP X
HK$150
(US$25)
Geotechnical Area Studies Programme - South Lantau (1988),
148 p. plus 4 maps.
GASP XI
HK$150
(US$25)
Geotechnical Area Studies Programme - Territory of Hong
Kong ( 1 989), 346 p. plus 14 maps.
GASPXH
HK$150
(US$25)
HB
'£
Geology of Sha Tin, by R. Addison (1986), 85 p.
Geological
Memoir No. 1
HK$50
(US$9)
Geology of Hong Kong Island and Kowloon, by P.J. Strange &
R.Shaw (1986), 134 p.
Geological
Memoir No. 2
HK$78
(US$12.5)
Geology of the Western New Territories, by R.L. Langford,
K.W. Lai, R.S. Arthurton & R. Shaw (1989), 140 p.
Geological
Memoir No. 3
HK$97
(US$17)
Geology of Sai Kung and Clear Water Bay by P.J. Strange, R.
Shaw & R. Addison (1990), 111 p.
Geological
Memoir No. 4
HK$87
(US$13)
Geology of the North Eastern New Territories, K.W. Lai,
S.D.G. Campbell & R. Shaw (1996), 144 p.
Geological
Memoir No. 5
HK$98
(US$17.5)
Geology of Lantau District by R.L. Langford, J.W.C. James, R.
Shaw, S.D.G. Campbell, P.A. Kirk & RJ. Sewell (1995),
173 p.
Geological
Memoir No. 6
HK$136
(US$28.2)
Geology of Yuen Long by D.V. Frost (1992), 69 p.
Sheet Report
No. 1
Free
Geology of Chek Lap Kok by R.L. Langford (1994), 61p.
Sheet Report
No. 2
Free
Geology of Tsing Yi by R.J. Sewell & J. A. Fyfe (1995), 43p.
Sheet Report
No. 3
Free
Geology of North Lantau Island and Ma Wan by R. J. Sewell
& J.W.C. James (1995), 46p.
Sheet Report
No. 4
Free
Geology of Ma On Shan by RJ. Sewell (1996), 45p.
Sheet Report
No. 5
Free
Geological Landscapes of Hong Kong (1998), 61p. (Bilingual).
HK$130
' ei
San Tin : Solid and Superficial Geology (1:20000 map)
(1989), 1 map.
Map HGM 20,
Sheet 2
HK$80
Sheung Shui: Solid and Superficial Geology (1:20 000 map)
(1992), 1 map.
Map HGM 20,
Sheet 3
HK$80
Kat O Chau : Solid and Superficial Geology (1:20 000 map)
(1993), 1 map.
Map HGM 20,
Sheet 4
HK$80
Tsing Shan (Castle Peak) : Solid and Superficial Geology
(1:20 000 map) (1988), 1 map.
Map HGM 20,
Sheet 5
HK$80
Yuen Long : Solid and Superficial Geology (1:20000 map)
(1988), 1 map.
Map HGM 20,
Sheet 6
HK$80
Sha Tin : Solid and Superficial Geology (1:20000 map)
(1986), 1 map.
Map HGM 20,
Sheet 7
HK$80
Sai Kung Peninsula : Solid and Superficial Geology (1:20 000
map) (1989), 1 map.
Map HGM 20,
Sheet 8
HK$80
Tung Chung : Solid and Superficial Geology (1:20 000 map)
(1994), 1 map.
Map HGM 20,
Sheet 9
HK$80
Silver Mine Bay : Solid and Superficial Geology (1:20000
map) (1992), 1 map.
Map HGM 20,
Sheet 10
HK$80
Hong Kong and Kowloon : Solid and Superficial Geology
(1:20 000 map) (1986), 1 map.
Map HGM 20,
Sheet 1 1
HK$80
Clear Water Bay : Solid and Superficial Geology (1:20000
map) (1989), 1 map.
Map HGM 20,
Sheet 12
HK$80
Shek Pik : Solid and Superficial Geology (1:20000 map)
(1995), 1 map.
Map HGM 20,
Sheet 13
HK$80
Cheung Chau : Solid and Superficial Geology (1:20 000 map)
(1995), 1 map.
Map HGM 20,
Sheet 14
HK$80
Hong Kong South and Lamma Island : Solid and Superficial
Geology (1:20 000 map) (1987), 1 map.
Map HGM 20,
Sheet 15
HK$80
Waglan Island: Solid and Superficial Geology (1:20 000 map)
(1989), 1 map.
Map HGM 20,
Sheet 16
HK$80
San Tin : Solid Geology (1 : 20 000 map) (1994), 1 map.
Map HGM20S
HK$80
Lo Wu : Superficial Geology (1:5 000 map) (1990), 1 map.
Map HOP 5A,
Sheet 2-NE-D
HK$100
Lo Wu : Solid Geology (1:5 000 map) (1990), 1 map.
Map HOP 5B,
Sheet 2-NE-D
HK$100
Deep Bay: Superficial Geology (1:5 000 map) (1989), 1 map.
Map HOP 5A,
Sheet 2-SW-C
HK$ 100
Deep Bay: Solid Geology (1:5 000 map) (1989), 1 map.
Map HOP 5B,
Sheet 2-SW-C
HK$100
Shan Pui: Superficial Geology (1:5 000 map) (1989),! map.
Map HGP 5A,
Sheet 2-SW-D
HK$ 100
Shan Pui: Solid Geology (1:5 000 map) (1989), 1 map.
Map HGP 5B,
Sheet 2-SW-D
HK$100
Mai Po : Superficial Geology (1:5 000 map) (1990), 1 map.
Map HGP 5A,
Sheet 2-SE-A
HK$100
Mai Po : Solid Geology (1:5 000 map) (1990), 1 map.
Map HGP 5B,
Sheet 2-SE-A
HK$100
Lok Ma Chau : Superficial Geology (1:5000 map) (1990),
1 map.
Map HGP 5A,
Sheet 2-SE-B
HK$100
Lok Ma Chau: Solid Geology (1:5 000 map) (1990), 1 map.
Map HGP 5B,
Sheet 2-SE-B
HK$ 100
Man Kam To : Superficial Geology (1:5000 map) (1990),
1 map.
Map HGP 5A,
Sheet 3-NW-C
HK$100
Man Kam To : Solid Geology (1:5 000 map) (1990), 1 map.
Map HGP 5B,
Sheet 3-NW-C
HK$100
Tin Shui Wai : Superficial Geology (1:5 000 map) (1989),
1 map.
Map HGP 5A,
Sheet 6-NW-A
HK$100
Tin Shui Wai: Solid Geology (1:5 000 map) (1989), 1 map.
Map HGP 5B,
Sheet 6-NW-A
HK$100
Yuen Long: Superficial Geology (1:5 000 map) (1989), 1 map.
Map HGP 5 A,
Sheet 6-NW-B
HK$ 100
Yuen Long: Solid Geology (1:5 000 map) (1989), 1 map.
Map HGP 5B,
Sheet 6-NW-B
HK$100
Hung Shui Kiu : Superficial Geology (1:5 000 map) (1989),
1 map.
Map HGP 5A,
Sheet 6-NW-C
HK$100
Hung Shui Kiu : Solid Geology (1:5 000 map) (1989), 1 map.
Map HGP 5B,
Sheet 6-NW-C
HK$100
Muk Kiu Tau : Superficial Geology (1:5 000 map) (1990),
1 map.
Map HGP 5A,
Sheet 6-NW-D
HK$100
Muk Kiu Tau : Solid Geology (1:5 000 map) (1990), 1 map.
Map HOP 5B,
Sheet 6-NW-D
HK$ 100
Tsuen Wan (Part): Solid & Superficial Geology (1:5 000 map)
(1995), 1 map.
Map HGP 5,
Sheet 6-SE-D
HK$100
Ma On Shan : Solid Geology (1:5 000 map) (1996), 1 map.
Map HGP 5B,
Sheet 7-NE-D,
C(part)
HK$100
Chek Lap Kok : Solid & Superficial Geology (1:5 000 map)
(1993), 1 map.
Map HGP 5,
Sheet 9-NE-C/D
HK$100
Tung Chung Wan: Solid & Superficial Geology (1:5 000 map)
(1995), 1 map
Map HGP 5,
Sheet 9-SE-A
HK$100
Pok To Van : Solid & Superficial Geology (1:5000 map)
(1997), 1 map.
Map HGP 5,
Sheet 9-SE-B
HK$100
Lantau Peak : Solid & Superficial Geology (1:5 000 map)
(1996), 1 map.
Map HGP 5,
Sheet 9-SE-C
HK$100
Sunset Peak : Solid & Superficial Geology (1:5 000 map)
(1996), 1 map.
Map HGP 5,
Sheet 9-SE-D
HK$100
Yam O Wan : Solid & Superficial Geology (1:5 000 map)
(1995), 1 map.
Map HGP 5,
Sheet 10-NW-B
HK$100
Siu Ho : Solid & Superficial Geology (1:5 000 map) (1994),
1 map.
Map HGP 5,
Sheet 10-NW-C
HK$100
Chok Ko Wan (Penny's Bay) : Solid & Superficial Geology
(1:5 000 map) (1994), 1 map.
Map HGP 5,
Sheet 10-NW-D
HK$100
Ma Wan: Solid and Superficial Geology (1:5 000 map) (1994),
1 map.
Map HGP 5,
Sheet 10-NE-A
HK$ 100
Tsing Y i : Solid & Superfical Geology (1:5 000 map) (1995),
1 map.
Map HGP 5,
Sheet 10-NE-B/D
HK$100
Pa Tau Kwu : Solid and Superficial Geology (1:5 000 map)
(1994), 1 map.
Map HGP 5,
Sheet 10-NE-C
HK$100
Tai Ho: Solid and Superficial Geology (1:5 000 map) (1995),
1 map.
Map HGP 5,
Sheet 10-SW-A
HK$100
ORDERING INFORMATION IS GIVEN ON THE NEXT PAGE
Copies of GEO publications (except Sheet Reports, 1:5 000 maps and other reports which are free of charge) may be ordered by
writing to :
•
Publications Sales Office,
Information Services Department,
28th Floor, Sm On Centre,
188 Lockhart Road, Wan Chai,
Hong Kong.
Fax (852) 2598 7482
H
ISjSxt 1 885^
^^'L^SS
EfcJfrSflffllS
«K (852) 2598 7482
The Information Services Department will issue an invoice upon receipt of a written order.
'
'
In Hong Kong, publications may be directly purchased from :
ltf (JtJ •
Government Publications Centre,
Ground Floor, Low Block,
Queensway Government Offices,
66Queensway,
Hong Kong.
Fax (852) 2523 7195
ffX
ifcflff O^SItFS
(852) 2523 7195
Requests for copies of Geological Survey Sheet Reports and other reports which are free of charge should be directed to :
:
'
Chief Geotechnical Engineer/Special Projects,
Geotechnical Engineering Office,
Civil Engineering Department,
Civil Engineering Building,
101 Princess Margaret Road,
Homantin, Kowloon,
fl£R
Hong Kong.
Fax (852) 2714 0275
SifehJtRSC EH&3EJ1
(852) 2714 0275
1 :5 000 maps may be purchased from :
1:5 000
'
Map Publications Centre/HK,
Survey & Mapping Office,
Lands Department,
23th Floor,
North Point Government Offices,
333 Java Road, North Point,
Fax (852) 2521 8726
4fX (852) 2521 8726
Ml prices given in this List are for information only and may be changed without notice. The US$ prices shown are for overseas
orders and are inclusive of surface postage to anywhere in the world. An additional bank charge of HK$50 or US$6.50 is required
per cheque made in currencies other than Hong Kong dollars. Cheques, bank drafts or money orders must be made payable to
THE GOVERNMENT OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION.
>
•
si >
'
-
»
575 - $
Latest information on the list of GEO publications can be found at the website http://www.info.gov.hk/ced/pub.htm on the Internet.
Abstracts for these documents can also be found at the same website.
fti?fl^±;frXgj®?!JtJ@$$ ' â±*Xg*Wa®IBIiHhttp://www.mfo.gov.h^ced/pub.htm±|î[j o a
•
HK 551.307 164
Investigation of some selected
landslide incidents in 1997
(volume 1)
Date Due
1
>

geo report - The University of Hong Kong

Transcription

Similar documents

PDF

View Presentation - University of the Philippines Diliman

China 2006 July 7 - 18 Beijing, Xi`an, Shanghai and Hong Kong

Please contact our Global Event Associate, Thu Thao Nguyen: E

m03 - The University of Hong Kong

APPS GAMES WEBSITES

S-05

PRESTIGE HONG KONG June 2011 THE WELL

Case Study of Landslides in Kabylia Region, Algeria

Geological Society of Malaysia Bulletin 49