Landslide Risk Management (2005) - Oldrich Hungr

Erlich, M. Fontoura, S.A.B. & Sayao, A.S.F. (eds.)Landslides evaluation and stabilization; Proceedings ofthe 9th International Symposium on Landslides. London:A.A. Balkema Publishers.

Seattle Times, 2001. Transportation windfall for state,November 30. Seattle: Seattle Times.

Simunek, J. Huang, K. & van Genuchten, M.Th. 1998. TheHYDRUS code for simulating 1-dimensional movementof water, heat and multiple solutes in variably saturatedmedia, version 6.0. Research Report 144. Riverside: U.S.Salinity Laboratory.

Srivastava, R. & Yeh, T.-C. J. 1991. Analytical solutions forone-dimensional, transient infiltration toward the watertable in homogeneous and layered soils. Water ResourcesResearch 27: 753762.

Wieczorek, G.F. 1987. Effect of rainfall intensity and dura-tion on debris flows in central Santa Cruz Mountains,

California. In Costa, J.E. & Wieczorek, G.F. (eds.) DebrisFlows/Avalanches: Process, Recognition, and Mitigation.Reviews in Engineering Geology. Boulder: GeologicalSociety of America.

Wieczorek, G.F. Morgan, B.A. & Campbell, R.H. 2000.Debris-flow hazards in the Blue Ridge of Central Virginia.Environmental and Engineering Geosciences 6: 323.

Wilson, R.C. 1997. Operation of a landslide warning systemduring the California storm sequence of January andFebruary 1993. In Larson, R.A. & Slosson, J.E. (eds.)Reviews in Engineering Geology. Boulder, CO:Geological Society of America.

Wilson, R.C. & Wieczorek, G.F. 1995. Rainfall thresholdsfor the initiation of debris flows at La Honda, California.Environmental and Engineering Geoscience 1(1): 1127.

740

Copyright 2005 by Taylor & Francis Group

Landslide Risk Management Hungr, Fell, Couture & Eberhardt (eds) 2005 Taylor & Francis Group, London, ISBN 04 1538 043 X

741

1 INTRODUCTION

This paper is primarily concerned with the manage-ment of landslide hazard and risk with the aid of a net-work of real-time continuously monitored web-enabledlandslide field stations. The importance of an obser-vational approach is implicit in this strategy. Four suchstations are currently operating in Wollongong, in thestate of New South Wales, Australia. A wider networkof field stations is proposed for the study area andextending to other landslide sites outside of the studyarea. The key features of these field stations are out-lined here and some data from three of the stations arepresented and discussed. Utilising the data from suchstations requires the development of multi-agent strate-gies to facilitate the efficient dissemination of infor-mation and response. These implementations of theobservational approach are one component of a com-prehensive strategy for the management of landslidehazard and risk employed within the Wollongong area.

1.1 The Wollongong area

The city of Wollongong is nestled on a narrow coastalplain approximately 70 km south of Sydney in the state

of New South Wales (NSW), Australia as shown inFigure 1. Over the last 150 years of settlement the pop-ulation of the Wollongong area has increased to about200,000 people. The coastal plain is triangular in shapewith a coastal length of 45 km. The coastal plain is upto 17 km wide in the south and narrows sharplytowards the north, disappearing north of Thirroul. Thecoastal plain is bounded to the north, west and southby an erosional escarpment ranging in height from300 m up to 500 m

The escarpment consists of slopes with moderateto steep inclinations with several intermediate benchesand cliff lines. Spectacular cliffs of HawkesburySandstone (of Middle Triassic age) cap the escarp-ment and there is dense vegetation over most of theescarpment below these cliffs.

The main road link to Sydney is the F6 Freewaythat traverses the escarpment via Mount Ousley Road.There are several other road links from the coastalplain to the top of the escarpment such as at BulliPass (refer Fig. 1). Lawrence Hargrave Drive links thenorthern suburbs to the F6 freeway via the spectacu-lar near vertical 200 m high cliffs near Clifton,although one section of this road is currently closeddue to landsliding (Hendrickx, et al. 2005, in prep).

Towards real-time landslide risk management in an urban area

P.N. Flentje & R.N. ChowdhuryFaculty of Engineering, University of Wollongong, Wollongong, NSW, Australia

P. TobinWollongong City Council, Wollongong, NSW, Australia

V. BrizgaRoads and Traffic Authority of New South Wales, Wollongong, NSW, Australia

ABSTRACT: With an average annual rainfall in the range 1200 mm1800 mm, landslides in Wollongong aretriggered by periods of prolonged heavy rainfall. During such events, real-time pore water pressure, slopemovement and rainfall data can be extremely useful for risk assessment and emergency management. A networkof continuously monitored real-time landslide field stations has been established in Wollongong. This networkwill facilitate risk management operations, enhance our understanding of landslide triggering mechanisms andimprove quantitative assessment of landslide hazard. Four stations have been established and additional stationsare proposed. Data collection and management is fully automated yet manually accessible as required. Themajor benefit of this system is the accessibility and availability of fully automated graphical output via the webimmediately as the data is received from the field stations. The widespread accessibility of this information innear real-time has clear benefits for risk management and emergency response decisions.


742

The South Coast and the Unanderra to Mossvalerailway lines also traverse the escarpment slopes andcoastal plain and both provide important freight andpassenger services between Sydney, Wollongong and the surrounding areas.

Processes and mechanisms of slope failure arecontrolled in Wollongong by factors such as stratigra-phy and geotechnical strength parameters, hydrogeol-ogy, geomorphology, slope inclination and pore waterpressure. Prolonged and/or intense rainfall is typicallythe trigger for significant landsliding. The average

annual rainfall for Wollongong varies from 1200 mmon the coastal plain near the city centre up to 1800 malong the top of the escarpment.

1.2 Background

The landslide research team at the University ofWollongong (UoW) has carried out systematicresearch over the last twelve years with funding fromthe Australian Research Council (ARC) as well assignificant support from several external industry

Figure 1. Location plan for the city of Wollongong showing the four field stations discussed in this paper.

http://www.crcnetbase.com/action/showImage?doi=10.1201/9781439833711.ch9&iName=master.img-041.jpg&w=391&h=395

partners. These include the Wollongong City Council(WCC), the Rail Corporation (RC), GeoscienceAustralia (GA) and, more recently, the Roads andTraffic Authority (RTA). During this research a numberof aspects have been covered including the following:

1 The development of a comprehensive GIS-baseddatasets including boreholes and structural geology

2 The development of a comprehensive landslideinventory containing 570 landslide sites (Flentje1998, Chowdhury & Flentje 2002)

3 The development of comprehensive GIS-basedmaps of geology and landslides

4 Geological and geotechnical modelling, determin-istic and probabilistic studies

5 GIS-based and numerical modelling of the spatialvariability and recurrence of rainfall

6 Development of a strategic framework for assess-ment of landslide hazard and risk including quali-tative and quantitative methods and includingscope for both site-specific and area-specific stud-ies (Ko Ko 2001, Ko Ko et al. 2004)

7 Knowledge-based modelling, using data miningtechniques, facilitating the preparation of GIS-based susceptibility maps for specific landslidetypes for a whole region (Chowdhury et al. 2002).

2 ESTABLISHING A REAL-TIMEMONITORING STATION

Equipment used to monitor a landslide in real-timecurrently starts at approximately $18,000 AUS in 2004and this does not include installation costs. This isclearly an inexpensive means of monitoring any land-slide which poses a significant or moderate risk toinfrastructure and especially if there is a moderate oreven low risk to human safety. At present, four remotelyaccessible continuous monitoring stations have beenbuilt in Wollongong and selected data obtained fromthree of these stations are discussed in this paper.

The instruments used in the Wollongong applica-tions include In-Place-Inclinometers (IPIs) andvibrating wire piezometers (vwp) installed at depth inboreholes. Rainfall Pluviometers have been installedat all the field stations to record rainfall as it occurs(0.2 mm or 0.5 mm bucket tips).

The stations are all powered by small solar panelswhich charge 12 Volt 7.0 Ah sealed lead-acid batterieshoused in Campbell Scientific PS/12 Power Supply/regulator units. Tele-communications are performedby Wavecom WMOD2 digital cellular mobile phones.Datalogging and on-site data management is carriedout with Campbell Scientific CR10X data loggers.Slope Indicator has bundled these systems togetherand supplied them to the University of Wollongongtogether with the programming for the CR10X data

loggers. Slope Indicator staff have completed theCR10X programming incorporating our research-basedlandslide triggering rainfall thresholds and other trou-bleshooting as required.

A discussion concerning the installation of incli-nometer casing and vibrating wire piezometers isbeyond the scope of this paper. However, it is worthnoting that the IPI instrument itself is approximately44 cm from wheel to wheel centre and approximately38.2 mm in diameter. These dimensions highlight twoimportant points. Firstly, the length can be extendedby the addition of stainless steel tubing to whatevergauge length movement is to be monitored over. Inone instance during local trials a 3 m-gauge lengthwas used, as the precise depth interval of shearing wasnot known. As landslide shearing progressed over a1 m depth interval within this 3 m interval, inaccurateA axis values were recorded as the stainless steelgauge tubing flexed within what has turned out to bea poorly backfilled inclinometer casing which alsoflexed in undesirable (upslope) orientations, as dis-cussed below. Clearly, as the depth and style of shear-ing is confirmed, the inclinometer gauge lengthshould be reduced to an appropriate minimum length.Secondly, the 38.2 mm diameter occupies more thanhalf of the typical 70 mm OD casing (approximately58.5 mm ID) restricting the amount of displacementthat can occur with the instruments remaining acces-sible and serviceable from the ground surface. Widerdiameter 85 mm OD casing (73 mm ID) inclinometercasing allows greater movement before serviceabilityissues will be encountered.

2.1 Real-time data acquisition and datamanagement

Data management in real-time is an integral aspect of this monitoring and the key component of theUniversity of Wollongong system. The LoggerNet andMultiMon software from Campbell Scientific andSlope Indicator respectively enable remote access tothe field stations from an office based PC with amodem and a telephone connection.

The programmable CR10X data logger is the intel-ligent component of each field station. The data log-gers are setup to record data hourly and in lowrainfall/dry times download data to the office weekly.When rainfall intensity increases the frequency ofdata download is increased to daily and even up to 4hourly (at which time the datalogger also starts record-ing data at 5 minute intervals). The rainfall intensitythresholds to trigger the varied data logger responsesare coded in the software over antecedent intervalsspanning 6 hours up to 120 days.

The office based PC can contact the field stationsat any time and download data. Alarm conditions canbe set on the office PC software whereby colour coded

743


data boxes on the graphical displays change colour asthresholds are reached and or exceeded.

The real challenge is then to disseminate this datato geotechnical colleagues and other managers in atimely fashion. This innovative aspect of the strategyadopted by the landslide research team is discussed inSection 7 below.

3 DATA FROM REAL-TIME MONITORINGSTATION INSTALLED AT SITE 113

Site 113 is a 3 m deep slide category landslide havinga volume of approximately 25,000 m3 that wasselected as the first trial research site for several rea-sons as summarised below:

1 It is an active shallow slide category landslide thathas destroyed 5 houses and 1 school building dur-ing the last 50 years

2 It has been reactivated 14 times over a period of 50 years (highest known frequency of 570 land-slides in study area) and was therefore likely toproduce useful data in a short time frame

3 It exposes a school yard to landslide risk4 A geotechnical investigation is currently ongoing

at this site

5 There is existing inclinometer casing at this siteand the depth of sliding is relatively well known.

The instrumentation (one IPI, one vwp and one plu-viometer) was installed during February 2003 and thestation was fully commissioned on 22nd March 2003.

The continuous monitoring record shown in Figure 2highlights a number of important features. The rain-fall over the 2-year period is close to average and thisis supported by the approx 1200 mm annual cumula-tive totals for both years. Still the period has producedthree relatively minor rainfall events during May2003, April and October 2004, which triggered somelandslide movement at this site. The May 2003 eventis discussed below. The pore water pressure curvealso displays two important points. The vibrating wirepiezometer was installed on the 3rd February 2003and did not indicate reasonable pore water pressureuntil late June 2003, a period of 4 months after instal-lation. However, since that time the pore water pres-sure data has shown considerable daily variation andsuperimposed on this variation a slow and steady risein pressure. As this piezometer is installed at 3.9 mdepth the variability displayed is considered to bepartly related to atmospheric pressure variation.

The hourly continuous monitoring record for May2003 is shown in Figure 3 and this clearly shows thatthe landslide accelerated during the afternoon of the

744

Figure 2. Continuous monitoring history, landslide Site 113, 22nd March 2003 to 7 Jan 2005. Rainfall, pore water pressure,landslide cumulative displacement and rate of shear.


13th May, reached a peak velocity of 0.4 mm/25 hourson the 15th May and slowed to 0.022 mm/day by the25th May 2003.

The landslide then briefly accelerated again up to0.15 mm/25 hours in response to 53.4 mm of rain onthe 26th May and slowed essentially to zero on the30th May. However, as shown in Figure 2, the slidedid continue moving episodically at extremely slowrates until late August 2003.

4 DATA FROM REAL-TIME MONITORINGSTATION INSTALLED AT SITE 355

Site 355 is a deep-seated slow moving slide categorylandslide with a volume of approximately 35,000 m3.A comprehensive geotechnical investigation of thislandslide, carried out by the University of Wollongong,has shown that this landslide presents a high risk ofloss of life for adjacent residential dwellings, the

proximity of which are clearly visible in Figure 4.This assessment was an important factor that resultedin the construction of this continuous monitoring station. Three IPIs, two vibrating wire piezometers andone rainfall pluviometer were installed at the site.

The continuous monitoring record of Site 355 isshown as Figure 5. Manual inclinometer profilesrecorded from borehole GUOWsc02, as shown inFigures 4 and 6, have recently confirmed that thedepth of sliding in borehole 2 is between 4 and 5 mdepth.

Up until 2nd November 2004, the upper IPI 11567at this site spanned the interval 4 m to 7 m that encom-passed the base of the fill material and upper collu-vium whilst the two deeper IPIs spanned the base ofthe colluvium and colluvium-bedrock interface.

The manual profiles also display two prominent up-slope deflections in the casing over the 5 to 6 m and 9 to13 m depths. The inclinometer was installed followingdifficult drilling operations and partially grouted and

745

Figure 3. Continuous monitoring record Site 113 for the May to August period 2003.


then backfilled with sand. Hence the upslope deflec-tions are considered to be the result of poor annulussupport. Hence any deflections indicated by IPIs11568 and 11569 have been ignored.

This continuous dataset has highlighted severalimportant issues. Firstly, the IPIs provide data regard-ing the relative displacement of two locations (per IPIinstrument) in a borehole, as opposed to the full bore-hole profile obtained from running a manual incli-nometer. Hence, the style (or profile) of landslideshear displacement experienced in a borehole shouldbe known before an IPI can be strategically placed.

Secondly, apparent negative displacement of IPI11567 as shown in Figure 5, was a concern. However,a review of the manual profiles displayed in Figure 6highlights the cause of the problem. IPI 11567 includesthe IPI instrument itself, as mentioned above, a 0.45 mlong instrument with the bottom wheels at the base ofthe 4 m to 7 m monitoring interval with a stainlesssteel tube 2.55 m long above that at the top of which isaffixed the top wheels of the monitored depth at 4 m.Figure 6 clearly shows that IPI 11567 was indeed being

rotated uphill (indicated negative displacement)approximately 18 mm whilst the soft stainless steeltubing was flexing in the inclinometer casing andactually masking the real landslide shear displace-ment. This situation has been rectified after the latestmanual profile was recorded by removing all threeIPIs and simply reinstalling one IPI (IPI 11568 wasarbitrarily selected) at the 4 m to 5 m depth interval.Preliminary data from this revised installation hasalready exhibited 2 mm downslope displacement.Interestingly, the Inclinometer casing has now exhib-ited a total displacement of approximately 90 mmover this 4 m to 5 m interval and the IPI was difficultto reinstall. Consequently we have already commis-sioned the installation of a new borehole in which wewill tremmie grout in place the larger diameter 85 mminclinometer casing. A new vibrating wire piezometerwill also be installed to monitor pore water pressureat this shallow depth.

Notwithstanding these problems, the IPIs haveprovided excellent data. The IPI 11567 rate of sheardisplacement curve in Figure 3 clearly displays two

746

Figure 4. Site 355 landslide. Continuously recording field station is located at site of borehole GUOWSc02. Crosses indi-cate the array of vector pegs.


prominent spikes of accelerated displacement com-mencing on the 4th April and the 21st October 2004.The movement event, which commenced on The 4thApril, continued for 5 days and peaked at 2.4 mm perday. This was triggered by rainfall of 110 mm and106 mm on consecutive days.

The movement event, which commenced on the21st October 2004, lasted for 7 days and peaked at2.7 mm per day (although this may have actually beenalmost 4 times higher at around 10 mm per daybecause of the problems discussed above) on the sec-ond day. A maximum daily rainfall of 81.5 mm andseveral other days of 15 mm to 30 mm triggered thisshort duration of movement.

The vibrating wire piezometers (vwp) have bothshown quite strong responses to the rainfall. However,it is important to note that these two instruments wereinstalled immediately above and below the bedrockcolluvium interface, 7 m and 9 m below the currentslide plane. As noted previously a new vwp will beinstalled at the depth of the slip surface in the nearfuture.

747

Figure 5. Continuous monitoring history, landslide Site 355, 11th October 2003 to 2 Dec 2004. Rainfall, pore water pressure,landslide cumulative displacement, rate of shear, vector Peg displacement and manual inclinometer displacement at 4 m depth.

Figure 6. Manual inclinometer profiles, boreholeGUOWSc02, Site 355.

http://www.crcnetbase.com/action/showImage?doi=10.1201/9781439833711.ch9&iName=master.img-045.jpg&w=393&h=295http://www.crcnetbase.com/action/showImage?doi=10.1201/9781439833711.ch9&iName=master.img-046.jpg&w=146&h=196

5 SITE 141 INTRODUCTION

Site 141 has been installed on a landslide that haspreviously disrupted three lanes of the six lane dualcarriageway of the F6 Sydney to Wollongong Freeway(Fig. 7). This field station is one of two that has beeninstalled on this road in collaboration with theSouthern Region of the Roads and Traffic Authority(RTA) of New South Wales during 2004. The secondstation is Site 153, 1.5 km south of Site 141 and as nomovement has occurred at this since the installationdate, data from Site 153 is not being discussed in this paper.

The Site 141 landslide is approximately 200 mwide, 60 m long downslope and up to 22 m of collu-vium has been encountered in boreholes drilled on thesite. These dimensions indicate an approximate vol-ume of 150,000 m3.

During 1988 a series of twenty 100 mm diametervertical eductor wells were installed at this landslidesite on the uphill side of the carriageway as shown inFigure 7 in an attempt to de-water and therefore,stabilize the landslide area. Each borehole was drilledto a depth of 2530 m and at 10 m centers spanned a200 m length of the road. Electrical pumps have beeninstalled in each eductor well and the outlet of each isinterconnected so that the daily volume of pumpedwater is logged.

Existing inclinometer records from this site span-ning 15 years since the installation of the eductorwells however, show an average annual displacementat this site of approximately 10 mm, with more move-ment in wet years and less in drier years.

The RTA is collaborating with the University andhas funded the stations as part of a strategy to; (a)gain a better understanding of how both sites are per-forming, (b) examine means of reducing maintenancecosts and (c) improve methods of managing risk toinfrastructure and life.

5.1 Data from real-time monitoring stationinstalled at Site 141

The two field stations on Mount Ousley Road werefully commissioned at 10:00am on Friday 30th July2004 and since that time negligible movement hasbeen detected as shown in Figure 8. However, tworainfall events, 76.2 mm and 164.8 mm on the 2ndand 22nd October 2004 respectively have resulted insignificant rises in pore water pressure at Site 141.Two vibrating wire piezometers have been installed atthis site in the same borehole, one at 17.9 m depthwithin the colluvium material, and the second at22.5 m depth within the bedrock material with exist-ing manual inclinometer profiles showing shear dis-placement is occurring between these two depths.

The rise in pore water pressure on the 2nd Octoberand the subsequent 3 days was striking and some causefor concern. However, the rapid drop in pore pressureon the 5th October was even more surprising. Thisprocess was repeated following the heavy rainfall ofthe 22nd October although this time it was accompa-nied by curious pore water pressure fluctuations.

The field station does not log the eductor well flowrate at present (may in the future), rather this data isacquired at 6 monthly intervals via data exchangewith another organisation. It is only upon viewing thisdata together with the pore water and rainfall data thata better understanding of the inter-relationships devel-ops. On the 12th July 2004, prior to the field stationbeing commissioned, the eductor wells pumping flowrate had dropped below the pre-set threshold of 90Kilolitres per day whereby the pumps were turned off,being re-activated again on the 4th October, followingthe rainfall on the 2nd October.

At this early stage it is clear that pumping from theeductor wells inhibits the rise in pore water pressureat both vibrating wire instrument levels (i.e., bothabove and below the slide plane) and that the porewater pressure rises that do occur are dissipated rap-idly. This of course has the positive effect of at leastmitigating landslide movement at this site.

6 DISCUSSION PROPOSED ARRAY OFSTATIONS DISTRIBUTED THROUGHOUTTHE STUDY AREA

There are four field stations operating in Wollongongat the present time and these represent the early stages

748

Figure 7. Site 141 on Mount Ousley Road (see Fig. 1).


of the University of Wollongong proposed network ofreal-time continuous monitoring stations. We proposeto install up to 15 additional stations at some of thenearly 600 landslides documented within the area thatpose a risk to residential areas or certain infrastruc-ture components.

The need for such an array of stations has been welldemonstrated during the August 1998 Wollongongrainfall event. During this event the city experienced750 mm of rainfall during 5 days and the city was iso-lated from adjacent urban centres including Sydneyfor 24 hours. A total of 142 landslides were activatedduring and in the weeks following this event. However,during the emergency response phase, accurate infor-mation regarding rainfall was limited and informationregarding landslide movement was not available,other than incoming reports of damage.

The proposed network of stations will facilitate theavailability of accurate information in real-time espe-cially during emergency management situations andthereby enhance the rational allocation of limitedresources during these peak demand situations.

The stations provide real-time information regard-ing the onset of landslide movement that is particu-larly important because of the episodic slip andstick nature of many of the Wollongong landslides.This is well demonstrated by the performance of Site355 during the recent late October 2004 rainfall eventas discussed above. The landslide accelerated on the

21st of October and the writers were aware at 9.00amon the morning of the 21st that this was occurring.Local government authorities were informed immedi-ately and inspections were carried out that morning.Over the next 7 days the event was monitored contin-uously and updated information was provided severaltimes a day.

7 WEB BASED MANAGEMENT OFCONTINUOUS REAL-TIME DATA

Having fully automated the data collection process atthe landslide sites and the transfer of that data to theoffice, the World Wide Web was considered to be themost appropriate way of managing the inbound dataand its dissemination. This is especially importantgiven the desired audience for the data being geo-technical colleagues, managers of essential infra-structure, managers of emergency services, policeand their technical advisers and in some cases, otherstakeholders.

Using the web to manage the data has importantbenefits. Firstly managing the data from the field sta-tions has proven to take considerable time in theoffice using the commercial software described pre-viously. Secondly, using the commercial software inthe office does not in itself get the essential data outto the required audience.

749

Figure 8. Continuous monitoring history, landslide Site 141 30th July to 7th Jan 2005. Daily and 6 hourly rainfall, porewater pressure, eductor wells daily volumetric output, landslide cumulative displacement and rate of shear.


Using the ASP.NET framework with a databasecreated in MS Access, the landslide research team incollaboration with the University of WollongongCentre for Educational Development and InteractiveResources (CEDIR) has developed web-based soft-ware to provide real-time graphical updates of theincoming data as it arrives from the field stations. Theweb-based facility is available via the University ofWollongong web portal http://landres.uow.edu.au/which opens as shown in Figure 9. At present four sitesare available and these can be selected from the menuon the left by clicking on the site locations on theindex map.

The site specific pages open as shown in the upperpart of Figure 10, in this case Site 355 has beenselected. The most recent 2 weeks of data is alwaysavailable at a glance by selecting the 2 week overview

button. Furthermore, the database of existing land-slide performance data is also available for review byselecting from a range of graphical outputs. The web-based hourly continuous monitoring record of InPlace Inclinometers displacement at Site 355 for the14 days up to 1st November 2004 12.00am is alsoshown in Figure 10. This 2 week period includes the21st23rd October movement event at this site.Graphs of hourly data displaying IPI Total displace-ment, IPI rate, IPI azimuth, hourly rainfall and porewater pressure for any 14 day period can be simplygenerated. The 5 graph types are soon to be extendedto 18 graph types and the display period is beingextended up to 180 days with a fully interactivegraphical user interface.

8 FUTURE DEVELOPMENTS

The web-based real-time facility is to be upgraded onseveral fronts during 2005. Firstly, due to the sensitivenature of the material presented the access will bepassword protected. At present the data supply fromthe field is based on a pre-programmed regional rain-fall intensity. As experience with the landslide sitesand instrumentation performance develops, thereporting of data could also be activated on the basisof specified magnitudes of (a) landslide displacement,or (b) rates of displacement, or (c) pore water pres-sure, or (d) refined site-specific rainfall thresholds.

The web-based software will, with the appropriateexperience, also be configured to provide alerts basedon rainfall, pore water pressure and or displacementthresholds. These alerts will be sent automatically todesignated staff via a range of media including emailand telephone (voice and text). Tabulated downloadsof data will also be enabled.

The network of field monitoring stations is also pro-posed to be extended beyond Wollongong to other land-slide areas within the state NSW and even interstate.

On a more general note the html address will alsobe expanded to access other areas of our landslideresearch at the Faculty of Engineering at the Universityof Wollongong.

9 CONCLUSIONS

Continuous real-time monitoring is an importantcomponent of quantitative landslide risk managementespecially during high magnitude (longer returnperiod) rainfall events and emergency managementoperations. However, it is also important for riskassessment work in helping to quantitatively assesslandslide frequency and hazard. In addition the dataprovides an important research component as thelandslide performance data together with pore water

750

Figure 9. University of Wollongong continuous real-timelandslide monitoring web page.

Figure 10. Site 355 continuous real-time monitoring webpage with IPI rate of displacement graph for the 2 weeks upto 12:00 am hours on the 1st November 2004.

http://www.crcnetbase.com/action/showImage?doi=10.1201/9781439833711.ch9&iName=master.img-049.jpg&w=190&h=114http://www.crcnetbase.com/action/showImage?doi=10.1201/9781439833711.ch9&iName=master.img-050.jpg&w=190&h=196

pressure and rainfall contributes greatly to the under-standing of landslide processes and triggering mech-anisms. Such data recorded at 5 minute and 1 hourintervals is providing an unparalleled database ofinformation.

Whilst Pedrozzi (2004) has recently suggested thatthe regional prediction of triggering of landslides isnot possible using rainfall intensity/frequency meth-ods in an area such as Canton Ticino in Switzerland,the writers strongly believe that a regional landslidetriggering rainfall threshold (intensity/frequency)curve is relevant for the Illawarra area of New SouthWales in Australia. In fact a preliminary threshold hasalready been proposed for this area (Flentje 1998,Flentje & Chowdhury 2001). It is understood thatrainfall threshold curves for specific landslide siteswill differ from a regional curve.

The continuous real-time monitoring discussed inthis paper will lead to a refinement of the existingregional landslide triggering rainfall threshold (intensity/frequency) curve and the refinement and ordevelopment of specific threshold curves for the foursites. These are important developments as they willenhance our ability to provide early warning of land-slide activity.

Coupled with these developments is the improvedquantitative assessment of landslide frequency andhazard that this data will provide. With experience asthe data record builds the rates of landslide displace-ment and frequency of events will be reviewed incomparison with structural damage and vulnerabilitytables.

ACKNOWLEDGEMENTS

The authors would like to acknowledge the supportand help of Slope Indicator staff in Australia Mr Colin Viska and in Vancouver Mr. Alan Jones.

REFERENCES

Chowdhury, R., Flentje, P., Hayne, M. & Gordon, D. 2002.Strategies for Quantitative Landslide Hazard Assessment.Proceedings of the International Conference on Instability Planning and Management. McInnes & Jakeways (eds.),Isle of Wight, UK, Thomas Telford, London, pp 219228.

Chowdhury, R. & Flentje, P. 2002b. Keynote Address -Modern Approaches for Assessment and Management ofUrban Landslides. Proceedings of the 3rd InternationalConference on Landslides, Slope Stability and the Safetyof Infrastructures. July 1112, Singapore. CI-PremierConference Organisation, pp 23 36.

Flentje, P. 1998. Computer Based Landslide Hazard and RiskAssessment. PhD Thesis, University of Wollongong,Australia.

Flentje, P. & Chowdhury, R.N. 2001. Aspects of RiskManagement for Rainfall Triggered Landsliding.Proceedings of the Engineering and Development inHazardous Terrain Symposium, New Zealand Geotechni-cal Society Inc. University of Canterbury, Christchurch,New Zealand. The Institution of Professional EngineersNew Zealand. August 2425, pp 143150.

Hendrickx, M., Wilson, R., Moon, A., Stewart I. & Flentje, P.in prep. Slope hazard assessment on a coast road in NewSouth Wales, Australia.

Ko Ko, C. 2001. Landslide Hazard and Risk AssessmentAlong a Railway Line, Ph.D. thesis, University ofWollongong, New South Wales, Australia

Ko Ko, C., Flentje, P. & Chowdhury, R. 2004. Interpretationof Probability of landsliding triggered by rainfall.Landslides 1(4): 263 275.

Pedrozzi, G. 2004. Triggering of landslides in canton Ticino(Switzerland) and prediction by the rainfall intensity andduration method. Bulletin of Engineering Geology andthe Environment 63(4): 281 291.

751


Landslide Risk Management (2005) - Oldrich Hungr

Documents

Transcript of Landslide Risk Management (2005) - Oldrich Hungr