Managing landslide hazards on the Illawarra escarpment

University of Wollongong Research Online Faculty of Engineering - Papers Faculty of Engineering 2005 Managing landslide hazards on the Illawarra escarpment. P. Flentje University of Wollongong, [email protected] R. N. Chowdhury University of Wollongong, [email protected] Publication Details This book chapter was originally published as Flentje, P and Chowdhury, RN, Managing landslide hazards on the Illawarra escarpment, in Morrison, J (ed), Proceedings of the GeoQuest Symposium on Planning for Natural Hazards - How can we mitigate the impacts? GeoQuesst Research Centre, University of Wollongong, 2005, 65-78. Research Online is the open access institutional repository for the University of Wollongong. For further information contact Manager Repository Services: [email protected]. MANAGING LANDSLIDE HAZARDS ON THE ILLAWARRA ESCARPMENT P.N.Flentje1 & R.N.Chowdhury2 Research Fellow1 and Emeritus Professor2 Faculty of Engineering University of Wollongong, NSW 2522, AUSTRALIA ABSTRACT ABSTRACT: For 12 year the University of Wollongong landslide research team has been developing a Geographic Information System based Landslide Inventory of the Wollongong Local Government Area and surrounding areas. This inventory can now provide data in near real-time via the World Wide Web. The inventory includes 569 landslide sites with a total of 956 landslide ‘events’ (includes all known occurrences and recurrences). Of these, four landslide sites are fully automated and continuously monitored, with data from them immediately accessible with a PIN access via the World Wide Web. Emergency response organisations and utility managers, as well as researchers and other stakeholders can monitor the movements, and this is especially important during times of high risk when direct access may be difficult. With an average annual rainfall across the city in the range 1200mm - 1600mm, landslides in Wollongong are primarily triggered by periods of prolonged heavy rainfall. The value to risk management of direct, near real-time access to data on landslide activity is obvious, and has been recognized world wide. The project will be extended by widening the area covered by the Inventory and near real time monitoring stations, and by refining the understanding of the rainfall thresholds likely to trigger landslide activity. INTRODUCTION This paper is primarily concerned with the management of landslide hazard and risks and incorporates a discussion of a comprehensive GIS-based Landslide Inventory and a network of continuous monitoring web-enabled landslide field stations. The landslide inventory has been developed for the Wollongong Local Government Area (LGA) and the surrounding areas. The landslide inventory is a key component of the landslide risk management strategy developed at the University of Wollongong for the Wollongong LGA. Projects where the landslide inventory has contributed are highlighted. The importance of an observational approach is implicit in this strategy. In addition to the Landslide Inventory, a network of real-time continuously monitored web-enabled landslide field stations is being developed. Four such stations are currently operating in Wollongong, in the state of New South Wales, Australia. A wider network of field stations is proposed for the study area and extending to other landslide sites outside of the study area. The key features of these field stations are outlined here and some data from three of the stations are presented and discussed. Utilising the data from such stations requires the development of strategies to facilitate the efficient dissemination of information and response. These implementations of the observational approach are one component of a comprehensive strategy for the management of landslide hazard and risk employed within the Wollongong area. The Wollongong Area The city of Wollongong is nestled on a narrow coastal plain approximately 70km south of Sydney in the state of New South Wales (NSW), Australia as shown in Figure 1. Over the last 150 years of settlement the population of the Wollongong area has increased to about 200,000 people. The coastal plain is triangular in shape with a coastal length of 45km. The coastal plain is up to 17km wide in the south and extends north to Thirroul. To the north of Thirroul, urban development exists on the lower slopes of the escarpment. The coastal plain is bounded to the north, west and south by an erosional escarpment ranging in height from 300 m up to 500 m Figure 1. Location plan for the City of Wollongong showing the four automated field stations. The escarpment consists of slopes with moderate to steep inclinations with several intermediate benches and cliff lines. The geological sequence encountered on the escarpment comprises an essentially flat-lying sequence of interlayered sandstone, mudstone and coal of the Illawarra Coal Measures, overlain by interbedded sandstones and mudstones/claystones of the Narrabeen Group. Spectacular cliffs of Hawkesbury Sandstone (of Middle Triassic age) cap the escarpment and there is dense vegetation over most of the escarpment below these cliffs. The main road link to Sydney is the F6 Freeway that traverses the escarpment via Mount Ousley Road. There are several other road links from the coastal plain to the top of the escarpment such as at Bulli Pass. Lawrence Hargrave Drive links the northern suburbs to the F6 freeway via the spectacular near vertical 200m high cliffs near Clifton, although one section of this road is currently closed due to landsliding (Hendrickx, et al., 2005). The South Coast railway line and the Unanderra to Moss Vale railway line also cross the escarpment slopes and coastal plain; both provide important freight and passenger services between Sydney, Wollongong and the surrounding areas. Processes and mechanisms of slope failure are controlled in Wollongong by factors such as stratigraphy, geotechnical strength parameters, hydrogeology, geomorphology, slope inclination and pore water pressure. The landslide inventory contains 3 main types of landslides, namely falls, flows and slides. Prolonged and/or intense rainfall is typically the trigger for significant landsliding. The average annual rainfall for Wollongong varies from 1200mm on the coastal plain near the city center and up to 1600mm along the top of the escarpment. Background The landslide research team at the University of Wollongong (UoW) has carried out systematic research over the last twelve years with funding from the Australian Research Council (ARC) as well as significant support from several key external industry partners. These include the Wollongong City Council (WCC), the Rail Corporation (RC), Geoscience Australia (GA) and, more recently, the Roads and Traffic Authority (RTA). During this research, a number of aspects have been covered: 1 The development of a comprehensive GIS-based datasets including boreholes and structural geology 2 The development of a comprehensive landslide inventory containing 569 landslide sites (Flentje, 1998, Chowdhury and Flentje 2002) 3 The development of comprehensive GIS-based maps of geology and landslides 4 Geological and geotechnical modelling, deterministic and probabilistic studies 5 GIS-based and numerical modelling of the spatial variability and recurrence of rainfall 6 Measurement of landslide movement with the aid of inclinometer monitoring and relating this movement to magnitudes of cumulative rainfall; and thereby identification of landslide-triggering rainfall thresholds 7 Development of a strategic framework for assessment of landslide hazard and risk including qualitative and quantitative methods and including scope for both sitespecific and area-specific studies (Ko Ko 2001, Ko Ko et al. 2004). 8 Knowledge-based modelling, using ‘data mining’ techniques, facilitating the preparation of GIS-based susceptibility maps for specific landslide types for a whole region (Chowdhury, R. et al. 2002). In 2004, the WCC and RTA both confirmed support for a further five-year University of Wollongong landslide research plan. In 2005, Rail Corporation with the support of the University of Wollongong won a Natural Disaster Mitigation Programme grant from the State Emergency Management Committee in collaboration with the Department of Transport and Regional Services (DOTARS) for a 5 year research project concerning landslide and flood hazard and risk management, with the aid of continuous and near real-time early warning systems along the South Coast rail corridor. In addition to the above, the Geelong City Council in the state of Victoria has also commissioned the installation of a University of Wollongong web-based continuously logging real-time landslide field monitoring station at an important landslide in that LGA. This field station is currently under construction and will be managed through the University of Wollongong web-based facility discussed below. GIS-BASED LANDSLIDE INVENTORY The GIS-based Landslide Inventory comprises digital landslide datasets (shapefiles in an ESRI ArcGIS Personal Geodatabase), from which maps are generated of all the known landslide sites. The Landslide Inventory has existed now for 10 years and has substantially grown in capacity every year since it was first developed (Chowdhury and Flentje 1998). The Inventory currently includes 569 landslide sites with a total of 956 landslide ‘events’ (including all known occurrences and recurrences). For example, Site 113 in Thirroul has 16 recurrences documented following its first recorded movement in March-April 1950. The key identifier for each record in the Landslide Inventory is the Site Reference Code, being a decimal number with one significant figure, is unique for each landslide site. An abbreviated data dictionary for the 22 standard fields required for each landslide site is shown in Table 1 and the database record ‘form’ showing these standard fields is shown in Figure 2. The database has a total of 75 fields available for each site and a comprehensive data dictionary is available the complete database. Table 1. Standard fields ‘data dictionary’ of the Landslide Inventory One aspect of the Landslide Inventory that has been extremely useful is the listing per site of the first occurrence and any subsequent recurrences of each landslide site (Flentje and Chowdhury, 2002). Such a listing for Site 113 is shown in Figure 3. This information is important for the assessment of landslide frequency, and provides significant evidence of landslide hazard. For example Site 113 was first reported in April to March 1950, and was most recently active in October 2004, a period spanning 54 years. With the 17 landslide events known at this site, the average annual frequency of landsliding is 0.315. With additional information regarding magnitudes or rates of displacement at each event, the frequency of landsliding can be defined even more precisely. Such calculations can directly be used in the quantitative assessment of risk. Figure 2. Database Record form showing the 22 standard fields for landslide Site 113. Figure 3. Landslide occurrence/recurrence data from Landslide Inventory for Site 113 In addition to the tabulated database, GIS-based maps of the known landslide locations can be prepared. An example of the mapping capability of GIS software is shown in Figure 4. GIS maps can be prepared at different scales, governed only by the resolution of the data displayed on the maps. One wall of the first writer’s office is covered with 1:10,000 scale maps of the Wollongong region containing, as a background, a 10m Digital Elevation Model of the region, cadastre, geology and superimposed on this landslides colour coded by landslide type. Figure 4. GIS-based map segment of Landslide Inventory showing the Scarborough and Wombarra areas in the northern suburbs of Wollongong. The University of Wollongong Landslide Inventory is now well known in the New South Wales geotechnical community. It is increasingly being used as a reference source for a range of infrastructure developments being considered or already in progress in Wollongong. The following list summarises the projects that have sought regional and or site-specific landslide data from the Landslide Inventory to date: • The Wollongong City Council and University of Wollongong landslide databases are combined to form one comprehensive Inventory • Roads and Traffic Authority of New South Wales Alliance partnership review of Slope Hazards affecting the Lawrence Hargrave Drive between Clifton-Coalcliff. This was part of the process for the $54 million bridge construction project which is now in progress • Rail Corporation of New South Wales review of landslide-triggering rainfall thresholds for the South Coast railway line • The New South Wales Department of Urban Affairs and Planning Commission of Inquiry into the long term planning and management of the Illawarra Escarpment and its foothills • Development of the Illawarra Escarpment Management Plan by the Wollongong City Council • National Parks and Wildlife Service of New South Wales exposure to landslide risk along the Wollongong escarpment undertaken by URS Pty Ltd • Assessment of the viability of a Rail Corporation of New South Wales realignment of South Coast railway line by Coffey Geosciences • Sydney Water Corporation development of Low Pressure Sewerage Scheme for the four Wollongong northern towns of Otford, Stanwell Park, Stanwell Tops and Coalcliff • University of Wollongong Landslide Research Team development of GIS-based landslide hazard maps for the WCC LGA using ‘data mining’ techniques. • Daily operations of the WCC related to geotechnical management of landslides within the LGA. • A variety of local and Sydney based geotechnical consultants daily operations related to management of landslides within the LGA and surrounding areas. • Input of all Wollongong landslide locations into the Australian Landslide database managed by Geoscience Australia. This list of projects clearly demonstrates the importance of the valuable information the Landslide Inventory contains. Having the information in one accessible location adds value to every project that accesses the information. The alternative of not having the accurate information accessible, regularly updated and in such a flexible format is unthinkable in the difficult and challenging Wollongong terrain. ESTABLISHING A REAL-TIME MONITORING STATION Equipment used to monitor a landslide in real-time currently starts at approximately $A20,000 in 2004, excluding installation costs. Such a level of expenditure is quite justifiable for monitoring any landslide that poses a significant or moderate risk to infrastructure, especially if there is a moderate or even low risk to human safety. At present, four remotely accessible continuous monitoring stations have been built in Wollongong, and selected data obtained from three of these stations are discussed in this paper. The instruments used in the Wollongong applications include In-Place-Inclinometers (IPIs) and vibrating wire piezometers (VWPs) installed at depth in boreholes. Rainfall Pluviometers have been installed at all the field stations to record rainfall as it occurs (0.2mm or 0.5mm bucket tips). The stations are all powered by small solar panels which charge 12 Volt 7.0 Ah sealed lead-acid batteries housed in Campbell Scientific PS/12 Power Supply/regulator units. Tele-communications are performed by Wavecom WMOD2 digital cellular mobile phones. Data-logging and on-site data management is carried out with Campbell Scientific CR10X data loggers. Durham Geo Slope Indicator (DGSI), a geotechnical instrument manufacturer, has bundled these systems together and supplied them to the University of Wollongong, together with the programming for the CR10X data loggers. DGSI staff have completed the CR10X programming incorporating our research-based landslide triggering rainfall thresholds and other troubleshooting as required. Real-time data acquisition and data management Data management in real-time is an integral aspect of this monitoring and a key component of the University of Wollongong system. The LoggerNet and MultiMon software, from Campbell Scientific and DGSI respectively, enable remote access to the field stations from an office-based PC with a modem and a telephone connection. The programmable CR10X data logger is the intelligent component of each field station. The data loggers are set up to record data hourly and in low rainfall/dry times download data to the office weekly. When rainfall intensity increases, the frequency of data download is increased to daily, and even up to 4-hourly (at which time the data logger also starts recording data at 5 minute intervals). The rainfall intensity thresholds to trigger the varied data logger responses are coded in the software over antecedent intervals spanning 6 hours up to 120 days. The office-based PC can contact the field stations at any time and download data. Alarm conditions can be set on the office PC software whereby colour coded data boxes on the graphical displays change colour as thresholds are reached and or exceeded. The real challenge is then to disseminate this data to geotechnical colleagues and other managers in a timely fashion. This innovative aspect of the strategy adopted by the landslide research team is discussed in the section headed web-based management below. DATA FROM SITE 113 REAL-TIME MONITORING STATION Site 113 is a 3m deep slide-category landslide having a volume of approximately 25,000m3 that was selected as the first trial research site for several reasons as summarised below: • It is an active shallow slide-category landslide that has destroyed 5 houses and 1 school building during the last 50 years • It has been reactivated 17 times over a period of 54 years indicating an average annual frequency of 0.315 (highest known frequency amongst 569 landslides in study area) and was therefore likely to produce useful data in a short time frame • It exposes a school yard to landslide risk • A geotechnical investigation is currently ongoing at this site • There is existing inclinometer casing at this site and the depth of sliding is relatively well known The instrumentation (one IPI, one VWP and one pluviometer) was installed during February 2003 and the station was fully commissioned on 22nd March 2003. The continuous monitoring record shown in Figure 5 highlights a number of important features. The rainfall over the 2-year period is close to average, with approximately 1200mm annual cumulative totals for both years. Even so, three relatively minor rainfall events that occurred during May 2003, April and October 2004, triggered some landslide movement at this site. The May 2003 event is discussed below. The pore water pressure curve also displays two important points. The vibrating wire piezometer was installed on the 3rd February 2003 and did not indicate ‘reasonable’ pore water pressure until late June 2003, a period of 4 months after installation. However, since that time the pore water pressure data has shown considerable daily variation superimposed on an increasing trend. As this piezometer is installed at 3.9m depth, the variability displayed is considered to be partly related to atmospheric pressure variation. Figure 5. Continuous monitoring history, landslide Site 113 22nd March 2003 to 7 Jan 2005. Rainfall, pore water pressure, landslide cumulative displacement and rate of shear. The hourly continuous monitoring record for May 2003 is shown in Figure 6 and this clearly shows that the landslide accelerated during the afternoon of the 13th May, reached a peak velocity of 0.4mm/25hours on the 15th May and slowed to 0.022 mm/day by the 25th May 2003. The landslide then briefly accelerated again up to 0.15mm/25hours in response to 53.4mm of rain on the 26th May and slowed essentially to zero on the 30th May. However, as shown in Figure 5, the slide did continue moving episodically at extremely slow rates until late August 2003. NETWORKED ARRAY OF STATIONS THROUGHOUT THE STUDY AREA There are four field stations operating in Wollongong at the present time and these represent the early stages of the University of Wollongong proposed network of realtime continuous monitoring stations. We propose to install up to 15 additional stations at some of the 569 landslides within the area. The need for such an array of stations has been well demonstrated during the August 1998 Wollongong rainfall event. During this event the city experienced 750mm of rainfall during 5 days and the city was isolated from adjacent urban centers including Sydney for 24 hours. A total of 142 landslides were activated during and in the weeks following this event. However, during the emergency response phase, accurate information regarding rainfall was limited and information regarding landslide movement was not available, other than isolated reports of damage. The proposed network of stations will facilitate the availability of accurate information in real-time, especially during emergency management situations, and thereby enhance the rational allocation of limited resources during these peak demand times. Figure 6. Continuous Monitoring record Site 113 for the May to August period 2003. The stations provide real-time information regarding the onset of landslide movement that is particularly important because of the episodic ‘slip and stick’ nature of many of the Wollongong landslides. This is well demonstrated by the performance of Site 355 at Scarborough (see Figure 4) during the late October 2004 rainfall event. The landslide accelerated on the 21st of October and the writers were aware at 9.00am on the morning of the 21st that this was occurring. Local government authorities were informed immediately and inspections were carried out that morning. Over the next 7 days, the event was monitored continuously, and updated information was provided several times a day. WEB BASED MANAGEMENT OF CONTINUOUS REAL-TIME DATA Having fully automated the data collection process at the landslide sites and the transfer of that data to the office, the World Wide Web was considered to be the most appropriate way of managing the inbound data and its dissemination. This is especially important given the desired audience for the data, i.e. geotechnical colleagues, managers of essential infrastructure, managers of emergency services, police and their technical advisers and in some cases, other stakeholders. Using the web to manage the data has important benefits. Firstly, managing the data from the field stations has proven to take considerable time in the office using the commercial software described previously. Secondly, using the commercial software in the office does not in itself get the essential data out to the required audience. Using the ASP.NET framework with a database created in Microsoft Access, the landslide research team in collaboration with the University of Wollongong Centre for Educational Development and Interactive Resources (CEDIR) has developed web-based software to provide real-time graphical updates of the incoming data as it arrives from the field stations. The web-based facility is available via the University of Wollongong web portal http://landres.uow.edu.au/ls/index.html which opens as shown in Figure 7. At present four sites are available and these can be selected from the menu on the left by clicking on the site locations on the index map. Figure 7. University of Wollongong Continuous Real-Time Landslide Monitoring Web page. The site-specific pages open as shown in the upper part of Figure 8. In this case, Site 355 has been selected. The most recent 2 weeks of data is always available at a glance by selecting the 2 week overview button. Furthermore, the database of existing landslide performance data is also available for review by selecting from a range of graphical outputs. The web-based hourly continuous monitoring record of In Place Inclinometer displacement at Site 355 for the 14 days up to 1st November 2004 12.00am is also shown in Figure 8. This 2 week period includes the 21st –23rd October movement event at this site. Graphs of hourly data displaying IPI total displacement, IPI rate, IPI azimuth, hourly rainfall and pore water pressure for any 14 day period can be simply generated. The 5 graph types are soon to be extended to 18 graph types and the display period is being extended up to 180 days with a fully interactive graphical user interface. FUTURE DEVELOPMENTS The web-based real-time facility is to be upgraded on several fronts during 2005. Firstly, due to the sensitive nature of the material presented, access to the web facility will be password protected. At present the data supply from the field is based on preprogrammed regional rainfall intensity. As experience with the landslide sites and instrumentation performance develops, the reporting of data could also be activated on the basis of specified magnitudes of (a) landslide displacement, or (b) rates of displacement, or (c) pore water pressure, or (d) refined site-specific rainfall thresholds. The web-based software will, with the appropriate experience, also be configured to provide warnings based on rainfall, pore water pressure and or displacement thresholds. These warnings will be sent automatically to designated staff via a range of media including email and telephone (voice and text). Tabulated downloads of data will also be enabled. The network of field monitoring stations is also being extended beyond Wollongong to other landslide areas within the state NSW and even interstate. As has been mentioned above, a station is currently being developed near Geelong in Victoria. On a more general note the html address will also be expanded to access other areas of our landslide research at University of Wollongong. Figure 8. Site 355 Continuous Real-Time Monitoring web page with IPI Rate of Displacement Graph for the 2 weeks up to 12:00am hours on the 1st November 2004 CONCLUSIONS The development of a landslide inventory of a region is a fundamental task and is a most effective means of gaining an understanding of the relevant issues affecting a given area. Such inventories should be widely accessible for land management purposes and be well maintained in the long term. Continuous real-time monitoring provides an important tool for landslide risk management especially during high magnitude (longer return period) rainfall events and emergency management operations. However, it is also important for risk assessment work in helping to quantitatively assess landslide frequency and hazard. In addition the data provides an important research component as the landslide performance data together with pore water pressure and rainfall contributes greatly to the understanding of landslide processes and triggering mechanisms. Such data recorded at 5 minute and 1 hour intervals is providing an unparalleled database of information. Pedrozzi (2004), working in Canton Ticino in Switzerland has recently suggested that the regional prediction of triggering of landslides is not possible using rainfall intensity/frequency methods. Quite clearly, site-specific information and data are required at key sites in addition to regional thresholds. The work of the University of Wollongong landslide research team has demonstrated that a regional landslide triggering rainfall threshold (intensity/frequency) curve is relevant for the Illawarra area of New South Wales in Australia. In fact a preliminary threshold has already been proposed for this area (Flentje 1998, Flentje and Chowdhury 2001). It is understood that rainfall threshold curves for specific landslide sites will differ from a regional curve. The continuous real-time monitoring discussed in this paper will lead to a refinement of the existing regional landslide triggering rainfall threshold (intensity/frequency) curve and the refinement and/or development of specific threshold curves for the four sites. These are important developments, as they will enhance our ability to provide early warning of landslide activity. Coupled with these developments is the improved quantitative assessment of landslide frequency and hazard that this data will provide. With experience as the data record builds up, the rates of landslide displacement and frequency of events will be reviewed in comparison with structural damage and vulnerability tables. ACKNOWLEDGEMENTS The writers would like to acknowledge the support and help of DGSI staff in Australia Mr Colin Viska and in Vancouver Mr. Alan Jones. The authors would also like to thanks Mr Russ Pennel and Mr Dhamika Ruberu from the University of Wollongong Centre for Educational Development and Interactive Resources for their assistance in the development of the web-based software. REFERENCES Chowdhury, R. N. and Flentje, P.N. 1998. A landslide database for landslide hazard assessment. Proceedings of the Second International Conference on Environmental Management. February 10 -13. Wollongong Australia. Editor: Sivakumar, M. and Chowdhury, R. N. Pages 1229 - 1239. Elsevier London. Chowdhury, R. and Flentje, P., 2002. Keynote Address - Modern Approaches for Assessment and Management of Urban Landslides. Proceedings of the 3rd International Conference on Landslides, Slope Stability and the Safety of Infrastructures. July 11 – 12, Singapore. CI-Premier Conference Organisation, pp 23 – 36. Chowdhury, R, Flentje, P. and Hayne, M and Gordon, D., 2002. Strategies for Quantitative Landslide Hazard Assessment. Proceedings of the International Conference on Instability – Planning and Management. Conference, editors: RG McInnes and Jenny Jakeways, May 2002, Isle of Wight, UK, Thomas Telford, London, UK, pp 219-228. Flentje, P. (1998). Computer Based Landslide Hazard and Risk Assessment. PhD Thesis, University of Wollongong, Australia. Flentje, P. and Chowdhury, R. N, 2001. Aspects of Risk Management for Rainfall - Triggered Landsliding. Proceedings of the Engineering and Development in Hazardous Terrain Symposium, New Zealand Geotechnical Society Inc. University of Canterbury, Christchurch, New Zealand. The Institution of Professional Engineers New Zealand. August 24-25, pp 143-150. Flentje and Chowdhury, 2002. Frequency of Landsliding as Part of Risk Assessment. Australian Geomechanics News, Volume 37 Number 2, May, pages 157 - 167. Australian Geomechanics Society, Institution of Engineers, Australia. Hendrickx, M., Wilson, R., Moon, A., Stewart I. & Flentje, P. 2005. Slope hazard assessment on a coast road in New South Wales, Australia. Proceedings of the International Conference on Landslide Risk Management, Vancouver (in press). Ko Ko, C., 2001. Landslide Hazard and Risk Assessment Along a Railway Line, Ph.D. thesis, University of Wollongong, New South Wales, Australia Ko Ko, C., Flentje, P and Chowdhury, R., 2004. Interpretation of Probability of landsliding triggered by rainfall. Landslides, Volume 1, Number 4, December, pp. 263 – 275. Springer. Pedrozzi, G., 2004. Triggering of landslides in canton Ticino (Switzerland and prediction by the rainfall intensity and duration method. Bulletin of Engineering Geology and the Environment. Volume 63, Number 4, pp. 281 – 291. View publication stats