A coupled hydrology/slope kinematics model for developing early warning criteria in the Kalitlaga Landslide, Banjarnegara, Indonesia - repository civitas UGM

  

Environmental Science and Engineering

Environmental Science

  Series Editors Rod Allan Ulrich Förstner Wim Salomons

  For further volumes:

• ^• Fawu Wang Masakatsu Miyajima _{• •}

  Tonglu Li Wei Shan Teuku Faisal Fathani

  Editors

Progress of Geo-Disaster

Mitigation Technology in Asia 123

  Editors Fawu Wang Wei Shan Department of Geoscience Institute of Engineering Geology Explora Shimane University Northeast Forestry University Matsue Harbin Japan People’s Republic of China

  Masakatsu Miyajima Teuku Faisal Fathani Kanazawa University Department of Civil and Environmental Kanazawa Engineering Japan Universitas Gadjah Mada

  Yogyakarta Tonglu Li Indonesia Department of Geological Engineering

  Chang’an University Xi’an People’s Republic of China

ISSN 1431-6250

  ISBN 978-3-642-29107-4 (eBook) DOI 10.1007/978-3-642-29107-4 Springer Heidelberg New York Dordrecht London _Ó Library of Congress Control Number: 2012944387 Springer-Verlag Berlin Heidelberg 2013

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of

the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,

recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or

information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar

methodology now known or hereafter developed. Exempted from this legal reservation are brief

excerpts in connection with reviews or scholarly analysis or material supplied specifically for the

purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the

work. Duplication of this publication or parts thereof is permitted only under the provisions of

the Copyright Law of the Publisher’s location, in its current version, and permission for use must always

be obtained from Springer. Permissions for use may be obtained through RightsLink at the Copyright

Clearance Center. Violations are liable to prosecution under the respective Copyright Law.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this

publication does not imply, even in the absence of a specific statement, that such names are exempt

from the relevant protective laws and regulations and therefore free for general use.

While the advice and information in this book are believed to be true and accurate at the date of

publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for

any errors or omissions that may be made. The publisher makes no warranty, express or implied, with

respect to the material contained herein.

  Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)

  Preface

  Japan Sea Region Research Institute was established 9 years ago in 2003 with the sole objective of bringing geoscience and geotechnically based researchers toge- ther. The founding members shared the idea of having an international research meeting which would accommodate other research areas that are related to the field of geotechnical engineering. Target participants and contributors were Jap- anese and foreign researchers around Japan Sea. The first international symposium was held in December 2003, in Kanazawa University Japan, under the title: ‘‘First International Symposium on Mitigation of Geo-hazards in Areas around the Japan Sea’’. Though the opening ceremony was missed by some foreign researchers due to heavy snowfall in and around the venue city, the symposium recorded huge success as everyone, through lively discussions, shared views and opinions on current scientific problems, and the participants unanimously decided to make the symposium an annual event.

  With success established, there was need to expand the research area to accommodate studies in geohazard mitigation and prevention like seismology and earthquake, building and pipeline engineering, and the target area was expanded beyond the countries around Japan Sea to larger areas including Indonesia, Iran, and so on.

  Scientific research has no end. The more we probe and solve environmental problems, the more we discover more unsolved problems. This is why we must not relent in providing answers and solutions to pertinent geoenvironmental chal- lenges. In commemoration of the tenth annual symposium, we are very happy to present to you the results of our latest research works. On behalf of the first chairpersons of the Steering and Organizing Committee in 2003, I would like to thank everybody so far present at the symposium and those who have in one way or the other contributed to the success of this event.

  v vi Preface

  Japan, May 22, 2012 Masaru Kitaura Chairperson of the First Steering and

  Organizing Committees Professor emeritus of Kanazawa University Foreword

  This book presents up-to-date research results of geo-disaster mitigation tech- nologies in Asia. Most of the contributions in this paper are from the core members and participants of the International Symposium on Mitigation of Geo-disasters in Asia (MGDA) in the 10 years, with some keynote papers from notable researchers who are very active in this region.

  As rightly introduced by Prof. Emeritus M. Kitaura, the first International Symposium on Mitigation of Geo-disasters in Areas around Japan Sea was held in 2003 at Kanazawa University, Japan. This pioneer event drew the attention of researchers and engineers from China, Korea, and Japan who were then interested or working on disaster mitigation around the Japan Sea region. The second, third, and fourth symposiums were held in the Kanazawa University while the fifth and sixth symposiums were held in 2007 and 2008 in Xi’an and Kunming cities both in China. In 2009, the name of the symposium was changed to ‘‘The International Symposium on Mitigation of Geo-disasters in Asia’’ to include more countries.

  The seventh event was held in Harbin city of China. Then, the eighth and ninth symposiums were held in Vladivostok city of Russia in 2010, and Yogyakarta city of Indonesia in 2011, respectively.

  The Department of Geoscience in Shimane University, Japan got the chance to organize the tenth event of the symposium in 2012. To celebrate the tenth anni- versary of the symposium, we made a proposal to publish a book with Springer to show the progress we have made in the past 10 years in the geo-disaster mitigation technologies in Asia. The idea was approved by the current steering committee which is made up of Prof. M. Miyajima (Kanazawa University, Japan), Prof. T. L. Li (Chang’an University, China), Prof. W. Shan (Northeast Forestry University, China) and Associate Professor T. F. Fathani (Gadjah Mada University, Indonesia), and an editorial committee for the book was formed.

  For the past 10 years, the continent of Asia has experienced a lot of geohazards like earthquakes, typhoons, and volcanic eruptions which have caused serious disasters. The reader is therefore enjoined to see some contributions made by the core members as we try to improve our efforts in providing timely results and suggestions on hazard assessment and disaster prevention.

  vii viii Foreword

  It is very important to acknowledge that all scientific approaches and methods employed in the study of hazard analysis and risk management are relevant in disaster mitigation. In Asia, many research institutes and investigation centers have presented diverse research methods and results based on their specific areas of interest. Our core members also made important contributions to geo-disaster mitigation. For example, the Kanazawa University research group concentrated on earthquakes and seismology and made major progress in lifeline earthquake engineering; the Chang’an University research group has made enormous contributions on disasters associated with loess deposits; the Northeast Forestry University research group are working on landslide disasters in cold regions; the Gadjah Mada University research group concentrated on tsunami caused by earthquakes, lahar by volcano eruption and landslide warning; the Jilin University research group conducted many research works on geotechnical and civil engi- neering problems, while the Shimane University group are working on submarine to localized landslides through experimental approach to numerical analysis, and the complex earthquake/rainfall-triggered landslides. It is our belief that the results and conclusions presented in these research areas will be of great benefit to the readers of the book.

  We also admitted the limitation of our existing network formed in the annual events, and are trying to enlarge the network so that we can include the key researchers of geo-disaster in the Asian region. So, in addition to the core members in our network, we invited Dr. A. Strom from Russia, Dr. Z. Q. Yue from University of Hong Kong, Dr. K. F. Liu from National Taiwan University, Dr. R. K. Dahal from Tribhvuan University, and so on to contribute some very comprehensive research results for the book.

  We appreciate the financial support from Kanazawa University, Chang’an University, and Northeast Forestry University which made it possible for us to publish this book. The Tenth International Symposium on Mitigation of Geo- disasters in Asia is also supported by a joint research fund courtesy of Disaster Prevention Research Institute, Kyoto University, and a germinating research project for disaster prevention of Shimane University. The San’in Disaster Pre- vention Forum based on Shimane University fully supported this important event. Mr. Austin Chukwueloka Okeke, Mr. Fikri Faris, and Mr. Hufeng Yang of Graduate School of Shimane University assisted the editing committee in editing this book.

  To celebrate the tenth anniversary of this annual event, we are organizing a field trip from Sendai city to investigate the 2011 disaster triggered by tsunami in the coastal area of eastern Japan, and also, analyze the achievements of Japan in earthquake-proof building designs. We enjoin everybody to join us to Mt. Bandai to see a deep-seated landslide triggered by volcanic eruption, and then move to Hotaka town to see debris flow countermeasures and monitoring systems. The field excursion will end with a trip to Shimane Prefecture to see the peaceful coun- tryside and a world heritage center: the Iwami Silver Mountain, where participants will directly feel the local people’s keenness to harmoniously live with natural hazards. Foreword ix

  Finally, we would like to dedicate this book to the founding members of the symposium: Prof. M. Kitaura and Prof. M. Miyajima for their academic foresight, leadership ability, and courage in mitigating disasters in Asia. We also appreciate the continued participation of our core members who have contributed so much to the success of this professional body. We believe that in the nearest future, our network will be much stronger, so that our scientific findings will not only benefit Asia, but will also be usefully applied in global disaster mitigation.

  May 28, 2012 Fawu Wang

  On behalf of the Chairpersons of the Tenth Steering and Organizing Committees Contents

  Part I Comprehensive Study . . . .

   Masakatsu Miyajima and Akira Murata . . . . . . . . . . . . . . . . . . . .

   Alexander Strom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

  . . . . . . .

   Ko-Fei Liu, Chyan-Deng Jan, Ping Sien Lin and Hsin-Chi Li Tonglu Li, Peng Wang and Yu Xi . . . . . . . . . . . . . . . . . . . .

   Ranjan Kumar Dahal and Netra Prakash Bhandary Dwikorita Karnawati, Teuku Faisal Fathani, Wahyu Wilopo and Budi Andayani Jiin-Song Tsai, Lap-Loi Chung and Karl Gee-Yu Liu

  xi xii Contents

   . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

   Fawu Wang, Ping Sun, Lynn Highland, Hongshuai Liu, Laizhen Pei and Qiangong Cheng

  Part II Specified Teuku Faisal Fathani and Djoko Legono . . . . . . . . .

   Iman Satyarno Masaho Yoshida, Masakatsu Miyajima and Atsunori Numata Bo Song, Fuqiang Qi and Zhongmao He Arjang Sadeghi Wei Shan, Zhaoguang Hu, Hua Jiang, Ying Guo and Chunjiao Wang Contents xiii

   Hua Jiang, Zhaoguang Hu, Ying Guo, Chunjiao Wang and Wei Shan Ying Guo and Wei Shan Balmukunda Regmi Saeed Alighadr, Abdolhossein Fallahi, Junji Kiyono and Masakatsu Miyajima Su-Chin Chen, Ko-Fei Liu, Lien-Kuang Chen, Chun-Hung Wu, Fawu Wang and Shih-Chao Wei Tonglu Li, Jifei Zhao, Ping Li and Fawu Wang Wei Shan, Fawu Wang, Ying Guo, Yuying Sun, Atsuo Takeuchi and Chunjiao Wang Fikri Faris and Faisal Fathani Yasuhiro Mitani, Fawu Wang, Austin Chukwueloka Okeke and Wenhao Qi xiv Contents

   Tomokazu Sonoyama, Fawu Wang, Mitsuki Honda and Yohei Kuwada Hufeng Yang, Wenxing Jian, Fawu Wang, Fanhe Meng and Austin Chukwueloka Okeke Austin Chukwueloka Okeke, Fawu Wang, Tomokazu Sonoyama and Yasuhiro Mitani Min Zhang, Lei Nie and Shulin Dai Shulin Dai, Chenglei Liu, Yan Xu and Tingxin Yu Yichen Zhang, Lei Nie and Yanliang Wang Chen Chen, Wei Wang, Youhong Sun, Wei Guo, Xuanchen Yan, Hongyan Wang, Guijie Zhao, Jiwei Wen, Qian Fang and Xinpeng Liu Shuheng Sun, Hepeng Zhang and Qing Dong Linlin Zou and Lei Nie

  Contents xv

   Atsuo Takeuchi, Kensaku Matsumoto and Tsuyoshi Harasawa Abdulhakim G. H. Al-Kholidi, Sun Youhong and Sun Zhifeng Editors Biography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

  A Coupled Hydrology/Slope Kinematics Model for Developing Early Warning Criteria in the Kalitlaga Landslide, Banjarnegara, Indonesia

  Fikri Faris and Faisal Fathani Abstract The tropical climatic condition of Indonesia brings consequences of high activity of landslide in the area. Although, studies of landslide activity and rainfall as triggering factor have been introduced in many research papers, the implementation of the research on the regional area which has particular geo- logical and climate condition is significant. This study has potential meaning to develop early warning systems that may avoid damage and loss of human lives in the area. In this research, field monitoring data have been assessed in Kalitlaga Village, Banjarnegara Regency, Central Java Province to develop coupled hydrology/slope kinematics model of landslide. To simulate the groundwater levels in the landslide, a hydrological model with two series of tank represented the geological setting of the area is applied. Groundwater output from the model was used in visco-plastic kinematics model to simulate the velocity of landslide movement. Calibration and verification of the model on the recorded data of groundwater and displacement data showed good results. The peak velocity from the combined model is then classified based on rainfall record of the previous landslide which happened near the area. Velocity that considered being critical condition is used as additional precursor of early warning system of Kalitlaga landslide. Keywords Hydrological model

  Visco-plastic Tank model Early warning system

  F. Faris (&) Graduate Student of Interdisciplinary Graduate School of Science and Engineering, Shimane University, 1060 Nishikawatsu, Matsue 690-8504, Japan e-mail: farisonsky@yahoo.com

F. Faris

  F. Fathani Department of Civil and Environmental Engineering, Faculty of Engineering, Universitas Gadjah Mada, Jalan Grafika No. 2, Yogyakarta 55281, Indonesia 453 F. Wang et al. (eds.), Progress of Geo-Disaster Mitigation Technology in Asia, Environmental Science and Engineering, DOI: 10.1007/978-3-642-29107-4_26, Ó

  454 F. Faris and F. Fathani

  1 Introduction

  Field investigations and laboratory analysis have revealed that the implementation of early warning system in landslide studies is important in landslide disaster management and control. A reliable landslide early warning system directly associates with good warning criteria. Comprehensive study about landslide mechanism derives warning criteria which integrates the triggering factors and probable landslide events. In the focus of rainfall triggered landslides, hydrological factors and its influences on landslide stability should be assessed to improve a better understanding of particular landslide mechanism.

  To facilitate landslide assessment and early warning system development, preliminary investigation and real-time field monitoring equipment was set up in Kalitlaga Village, Banjarnegara District, Central Java Province of Indonesia as a pilot project (Fathani et al.

  

. The development of the project in the area was part of the Asian Joint Research

  Project for early warning system on landslides disaster management, which consists of the International Consortium on Landslides (ICL), Gadjah Mada University (GMU) and Disaster Prevention Research Institute of Kyoto University (DPRI/KU). As part of the project, real-time field monitoring equipment such as rain-gauges, groundwater level sensors and extensometers were employed to obtain data of precipitation, groundwater level and landslide mass displacement, respectively. These data are essential for the landslide assessment, prediction and early warning system development.

  2 Landslide Study Area

2.1 Kalitlaga Landslide

  The studied area is in Kalitlaga village which is located in Pagentan District, Banjar- negara Regency of the Central Java Province of Indonesia (Fig.

   ). The lithological

  composition of the area is characterized by Rambatan Formation which consists of shale, marl and calcareous sandstone (Condon et al. ). Specifically, the geological structure of this area was inclined stratigraphic of sediment layers which consist of montmorillonite clay covered by colluvial deposits (Karnawati et al. ).

  In view of the imminent landslide hazard in the area, a landslide monitoring system was established in order to assess, track, monitor and record the sequence and failure mechanism of the slope due to intermittent rainfall. The installed field monitoring equipment, as previously discussed, consist of outdoor unit and indoor unit. The outdoor unit, which is made up of strategically positioned sensors and electronic gadgets like long-span extensometers, rain-gauges, internet protocol (IP) cameras and water pressure sensors, records information observed from the field, while the indoor unit processes, stores, and sends real-time information via a GPRS modem which is displayed on a web-server (Fathani et al. For the further analysis concern, a reliable data of topography and aerial photo has been reserved through Balloon

  

A Coupled Hydrology/Slope Kinematics Model 455

Fig. 1 Location of Kalitlaga landslide

  Photogrammetry System which combines balloon aerial photography and digital photogrammetry for low-altitude aerial mapping. The balloon carries out a digital camera up to 400 m above the ground level and takes aerial photographs in appropriate viewing angle. The digital photogrammetry processes photo-restitution to produce 3D model from multi-view aerial photograph (Rokhmana ).

2.2 Hydrological Influence on Landslide Activity

  From hydrological point of view, groundwater is highly affected by rainfall activity and infiltration rate that controlled by land surface condition. Since there is a lot number of cracks and cultivated area comprise the landslide body, the rainfall can be easily infiltrate into the soil and become groundwater. Angeli et al. ) considered that the rapid piezometric response is attributed to an infiltration pro- cess through a superficial system of interconnected cracks. This system not only works as a fast flow-path from the ground surface to water table, but is also capable to discharge the infiltrated water rapidly.

  456 F. Faris and F. Fathani

  Fig. 2 The influence of 400 ) e groundwater flow against soil c a

  350 0.5 rf mass displacement u s

  300 mm)

  1 low groundwater level 250 t ( be m

  1.5 200 r ( displacement te lacemen a

  150

  2 isp D 100

  2.5 roundw rainfall

50 G

  3 1/1/09 4/1/09 7/1/09 10/1/09 1/1/10 Date

  The water pressure sensor attached in an observing well showed that the ground water fluctuation is highly affected by rainfall event. The observation by field monitoring also revealed that there is a high influence of groundwater fluctuation to soil mass displacement that is recorded in specific area (Fig.

  Based on this

  fact, a sound geotechnical model which appropriately predicts the soil mass dis- placement by considering hydrological factor is needed to be developed.

2.3 Hydrological Model

  In some studies, hydrological model used for estimating groundwater level con- siders the geological formation that occurs in particular area. Hong et al. ) have taken the three cascading tank into consideration based on the geological formation that is indicated from the investigation of borehole in Zentoku landslide on Shikoku Island, Japan. Angeli et al. ( also represented the incorporation of geological structure in hydrological model in the study of Alverà landslide, Italy. The developed model in Alverà consists of two reservoirs that represent the root zone and the clay layer. Each of studies introduced different hydrological model according to particular geological formation in the study area. Developed hydro- logical model in both studies that is previously described needs actual groundwater level data and rainfall data to verify the results.

  In this study, required data to support the development of the model are already obtained through pilot project that has been established in Kalitlaga landslide. The general geological condition in the site, such as geomorphology of the area, the lithol- ogy, geological structure and slope stratigraphy has been studied through geological investigation. The specific geological formation in the area arises some consideration of hydrological concept which is can be described schematically with tank model.

  The tank model consisted of two series of tank as shown in Fig.

   . This model

  was based on assumption that the first tank represented the water balance at the surface that was controlled by runoff and infiltration while the second tank rep- resented the colluvium layer. The first tank is used to simulate the surface run-off

  

A Coupled Hydrology/Slope Kinematics Model 457

Fig. 3 Tank model configuration in Kalitlaga landslide

  and infiltration into colluvium soil. Once the rainfall enters the first tank, then it is distributed to the lower outlet D

  1 , in case the previous water level at the first tank

  is still below h

  11 . Otherwise, if the water level at the first tank is same as or higher

  458 F. Faris and F. Fathani

  Table 1 Parameters of tank Parameters Value model s 100.0 b ¹

  0.4 a ¹¹

  0.03 a ²¹

  0.01 a ²²

  0.16 a ^{23 0.0025} h ¹¹

  1 h ²¹

  7 h ₂₂

  2 h ₂₃

  than h , then the rainfall becomes runoff Q . The height h corresponds to soil

  11

  1

  11

  moisture capacity. Once the water level at the first tank surpasses h , one can

  11

  interprets that the soil is already saturated and the next rainfall adding up will becomes surface runoff. The second tank represents the input and output of water at the borehole, thus the fluctuation of water level in the second tank corresponds to groundwater level. The second tank receives discharge from infiltration D that

  1

  subsequently distributed to three outlets of the second tank, which represent the disparity of soil permeability at the borehole.

  Land surface and geological condition highly control the infiltration and ground- water fluctuation. Since there is a lot number of cracks and cultivated area comprise the landslide body, the rainfall can be easily infiltrate into the soil and become ground- water. However, groundwater does not exist homogenously under the ground. Groundwater tends to congregate into such a stream due to the diversity of soil per- meability (Takeuchi ). The lateral flow from recharge area, which is affected by existence of cracks, asperity and concave-like shape of landslide base, is assumed to contribute on this phenomenon. Groundwater stream mechanism can be illustrated as ‘‘underground river’’ with some catchments area. However, neither the extent of this area, nor its geohydrological properties are known, and both entities are difficult to assess (Angeli et al. Although there is no inflow from upper slope that explicitly represented in tank model, the scale factor (s) is used to represents the inflow from catchments area surrounding the landslide. The scale factor (s) obtained by calibration and one can conclude it as the representation of catchments area broadness. This factor is applied as multiplier factor of total height of the last tank in tank model (H).

2.4 Result of Hydrological Model This simulation needs consistent and continuous rainfall data and groundwater level.

  Thus, rainfall event January 11, 2009–June 19, 2009 which is adequately supported by continuous groundwater level data considered as stored data. The heights, the coeffi- cients and the numbers of outlets at the last tank are determined by trial and error until the simulated groundwater level presents a good corresponding relation with the real

  

A Coupled Hydrology/Slope Kinematics Model 459

Date

  

1/1/09 2/1/09 3/1/09 4/1/09 5/1/09 6/1/09

100

  50 200 300

  100 400 150 500 600

  200 700

  Hourly rainfall (mm) Groundwater depth (mm)

  800 250 rainfall Simulated Groundwater level 900 Actual Groundwater level 300

  1000 Fig. 4 Simulation result of the groundwater level from tank model

  groundwater level. This relation can be expressed by the linear regression which results the simulated levels in boreholes in millimetres. Table

   shows adjusted coefficient values and height of the outlets to satisfy required condition.

  The verification used statistical parameter of coefficient of correlation (R) and volume error (V ) which are obtained from linear relationship of simulated ground-

  e

  water level with the actual groundwater from field monitoring. The result of simulation is shown in Fig.

  The red curve that displays the estimated hydrological response to

  the rainfall during the chosen intervals proves that the tank model could simulate the groundwater with a good precision. The graphic also shows high response of groundwater level due rainfall activity. The possible reasons of this phenomenon, possibly, are the occurrence of cracks and highly cultivated land which eases infil- tration. The obtained coefficients of correlation confirm the good corresponding

  ) of simulation result between these two parameters (R = 0.927). The volume error (V e against the observed data also gives permitted value (V e = 0.042 % \ 5 %).

2.5 Visco-Plastic Assumption for Slope Kinematics Model

  Slow slope movements are typically associated with ‘‘creep’’ behaviour, since the soil can be characterized by a viscous response, implying the presence of a resisting force that is activated only when the soil mass starts to move (Ranalli et al. ).

  For continuously moving landslides, a dynamic analysis should be adopted instead a classical static approach (Ranalli et al. ). A classical static analysis is suitable to determine the slope safety factor, but is not able to model the actual kinematics of the soil mass behaviour. In the limit equilibrium stability analysis, the soil shear strength is usually defined by the Mohr–Coulomb criterion, and the instability condition occurs when the equilibrium is modified by a pore water pressure increase and a consequent reduction of the effective stress level. For a given piezometric level, a constant instability force could exist and initiate a

  460 F. Faris and F. Fathani

  slope movement with constant acceleration and a corresponding velocity, linearly increasing with time. When these conditions fail, a possible explanation is the effect of a viscous resisting component of the material. In this case, the mass velocity can be related to the excessive shear stress by different viscous laws, like the Bingham’s law, which shows a yield point and a subsequent linear relationship (Ranalli et al. ): v ¼ 0 s\s ð1Þ v ¼ ðz=gÞðs s v Þ s s ð2Þ where v is the velocity (m/s), z is the shear band thickness (m), g is the dynamic

• _-2

  2

  2

  viscosity (N.s.m ), s is the driving shear stress (N/m ), and s (N/m ) is the resisting shear strength (yield strength). The resisting shear strength s can be defined by the Mohr–Coulomb failure criterion at the residual strength state (Van Asch and Van Genuchten This yield strength line can be described by the following equation: tan s ¼ c þ r / ð3Þ

  r r

  where: s = the yield strength; c = the residual cohesion; r = the effective

  r

  normal stress; and / = the residual friction angle.

  r

  Once the frictional resistance of weight is overcome it starts moving in creep phase with a velocity controlled by the viscous resistance of the dash-pot which is expressed by: s ¼ g v ð4Þ

  B

  where g corresponds to viscous characteristic of the soil in the dash-pot and v is

  B the velocity the sliding block (Angeli et al.

  The viscous parameter C is defined as the ratio between the material dynamic viscosity g and the thickness of the shear band z. The term C = g/z is called as the coefficient of dynamic viscosity. As a consequence, the contribution to resistance can be expressed as s = Cm which analogous to the Bingham’s expression for m. The C value obtained viscous characteristic of the oil in the dash-pot s = g B from the calibration process (Ranalli et al. ).

  The dynamics of the landslide are governed by the difference between driving forces (F), that depend basically on weight and slope, which are constant and resisting forces (F ) are sensitive to water pressure at the slip surface (Corominas

  r

  et al. The momentum equation can be written as: F F r ¼ ma ð5Þ _-2 where m is the mass (kg) and a the acceleration (m.s ). Resisting forces can be estimated using Mohr–Coulomb criterion which considers cohesion and friction parameter. By using infinite slope condition, forces are computed over a unit surface, and therefore shear stresses are considered in what follows; s c ½ þ r p ð Þ tan / ¼ ma ð6Þ

  

w

  

A Coupled Hydrology/Slope Kinematics Model 461

Fig. 5 Model of Bingham body

  where s is driving shear stress, c is the cohesion, r is the normal stress, p the pore

  w water pressure and / the friction angle, all magnitude projected to the slip surface.

  Figure is already presents the variables used in this approach.

  Corominas et al. (

   ). That is:

  s c ½ þ r p ð w Þ tan u s v ¼ ma ð7Þ Viscous forces are usually dependent on the strain rate of the shear zone. For a

  Bingham typemodel, this relation is linear and becomes: s ¼ gv=z ð8Þ

  v where, g is the viscosity, v is the velocity and z is the thickness of the shear zone.

  Expression

  to give differential equation for a

  single point, assumed representative of the whole landslide: dv v

  

2

  cl sin a cos a a c þ cl cos p w tan / ¼ m þ g ð9Þ dt z

  Pore water pressure was not measured directly, but it was estimated from readings of depth of groundwater level. Assuming a parallel flow to the slope surface:

  2

  2

  cos cos p ah a

  w ¼ c ¼ c ðl D w Þ ð10Þ w w

  where c is the specific weight of water, l the thickness of the sliding mass and D

  w w the depth of groundwater level.

  462 F. Faris and F. Fathani

  Fig. 6 Infinite slope with a viscous deforming layer (z)

  The governing equation which expressed in Eq. (

  is considered as first order

  ordinary differential equation. By means of integration, this equation can be solved analytically in terms of velocity by considering the following form: dy

  þ yPðxÞ ¼ QðxÞ ð11Þ dx In order to match up the governing equation to general form, the Eq. (

  need to

  be modified by dividing both two side with terms m. Hence, the Eq.

   ) can be

  rewritten as Eq.

   ) that is:

  dv þ B ¼ A ð12Þ dt where

  2

  cl sin a cos a a c ½ þ cl cos ð p w Þ tan u A ¼

  ð13Þ m and

  

A Coupled Hydrology/Slope Kinematics Model 463

Table 2 Soil parameter Parameter Value ₃ Saturated unit weight, c sat (kN/m )

  20.8 Effective internal friction angle, / (degree) ₂

  15.4 Effective cohesion, c (kN/m )

  10.4

  g B ¼ ð14Þ z m

  Through integration procedure, the Eq.

  yields the solved governing equation:

  A

  Bt

  v ¼ 1 e ð15Þ B

  3 Combined Hydrological and Slope Kinematics Modelling

  The simulated groundwater level from tank model were used as input to the kine- matics model to directly obtain displacement values as a function of precipitation. The soil parameters that used in kinematic model are obtained from laboratory test of index properties and CU triaxial test which is shown in Table

  The result of the

  displacement simulation of Wire 4–5 and Wire 5–6 for the period January 2009– January 2010 are shown in Figs.

   , respectively. The distinctive parameter

  between these two simulations is the coefficient of dynamic viscosity (C). This parameter determined through calibrations process until the simulated displacement curve and actual displacement curve are considered to be have a good fit. Wire 5–6 shows the most responsive and highest velocity among the extensometers, thus it is considered to be used for further development of early warning criteria.

  Obviously, the use of this model and its parameters are limited to local con- ditions and being a representative for particular area. However, it is believed that this area can be a sort of additional precursor of landslide warning system for the surrounding areas which have more susceptible against landslide prone, i.e. steeper slope and highly populated.

  4 Early Warning Criteria

  The importance of early warning system arises when other mitigation efforts in avoiding landslide and protecting landslide mass considered being costly and ineffective, meanwhile human lives threatened by landslide disaster anytime.

  A good early warning system should give an accurate prediction of landslide occurrences. This kind of ideal early warning system is built by comprehensive understanding of landslide mechanism and supported by representative and accurate data. Thus, the addition of supporting data may increase the accuracy of prediction.

  464 F. Faris and F. Fathani

  0.1 400 Velocity

  0.09 350 Simulated Displacement

  0.08 Actual Displacement 300 )

  0.07 /h mm) m

  250 t (

  0.06 n (m me

  0.05 200 e ity c c lo la

  0.04 e p

  150

  V is D

  0.03 100

  0.02

  50

  0.01

1/1/08 5/1/08 9/1/08 1/1/09 5/1/09 9/1/09 1/1/10

Date

  Fig. 7 Simulated velocity, simulated displacement and actual displacement of Wire 4–5

  0.25 350 Velocity 300 Simulated Displacement

  0.2 Actual Displacement 250

  0.15 200 150

0.1 Velocity (mm/h)

  100 Displacement (mm)

  0.05 50 1/1/08 4/1/08 7/1/08 10/1/08 1/1/09

  

Date

Fig. 8 Simulated velocity, simulated displacement and actual displacement of Wire 5–6

  Formerly, the developed model is able to generate landslide mass creep velocity based on real-time rainfall data obtained from the field. In this particular region of Kalitlaga landslide, where the real-time monitoring implemented, some high precipitation would not emerge a dangerous velocity since the landslide mass just tends to creep. However, in case of other steeper slopes at surrounding area, such kind of precipitation might be disastrous. Hence, it needs to correlate the landslide mass creep velocities in this monitored area with some real landslide event which has happened in vicinity area. Usually, several rainfall data need to be collected in order to support a reliable prediction. However, in this case, the only available data once landslide occurred in Kalitlaga on December 30th 2007 just near the cur- rently studied landslide location is daily rainfall intensity data. However, to develop early warning criteria, the determination of hourly rainfall intensity should be included. Thus, the hourly rainfall distribution is generated based two methods of rainfall distribution; Tadashi Tanimoto method and Alternating Block Method

  

A Coupled Hydrology/Slope Kinematics Model 465

Table 3 Landslide mass Velocity class Range (mm/h) velocity class _\ Class 1 (C1) _[

  0.15 Class 2 (C2)

  0.15 Fig. 9 SWI and hourly rainfall plots and recorded peak velocities from the model

  (ABM). Tadashi Tanimoto method based on his study about rainfall distribution in Java Islands, whereas ABM method using some formulas in determines the rainfall duration and distribution (Aras ).

  Subsequently, these rainfall data were stored in previously developed model to generate soil water index (SWI) of each rainfall event. SWI is defined as the total water level height of the tanks series (Okada In this study, a new proposed method to built early warning criteria is described by associating the Soil Water Index (SWI) and landslide mass velocity as additional precursor. The determina- tion of SWI from critical rainfall event is achieved initially and its further cor- relation with landslide mass velocity from the combined model taken into account afterward.

  The generated SWI from each rainfall scenario is plotted in X–Y graph. The critical line (CL) is then drawn by intersecting two scenarios which is expressed by line Y ¼ X þ 32. This line is considered as the safest based on available data.

  In order to build some additional precursor using landslide mass velocity from particular site, it is needed to divide the velocities into several categories/classes. The velocities are divided in two classes described in Table

   . The velocity range

  is categorized using line CL as demarcation line between velocity ranges. The plot of peak velocity is generated from developed combined simulation using recorded 2 years rainfall data from January 18, 2008 to January 18, 2010 is shown in Fig.

   .

  466 F. Faris and F. Fathani

5 Conclusions The combined model reliability is strongly depending on data recording validation.

  The reliable and continuous data leads to high performance of both hydrological model and kinematics model and their combination. Recently, this model is considered as the representative of particular landslide area. However, further study of comparable landslide on similar geological condition with better quan- tification and verification is expected to provide more understanding of landslide movement triggered by rainfall.

  The combined hydrological model and kinematics model showed a good agree- ment between computed and actual displacement using estimated parameter of tank model. The simulated velocity is possible to be correlated to rainfall parameter generated by hydrological model to develop early warning criteria. The determi- nation of early warning criteria by means of rainfall data analysis and soil water index (SWI) obtained from developed tank model gives a line form of critical line (CL) Y

  ¼ X þ 32 where x-axis and y-axis corresponds to SWI and hourly rainfall, respectively. In the mitigation point of view, landslide mass velocities obtained from the combined model result contributed in being additional precursor in early warning criteria by dividing landslide velocity range into two classes.

  References

Angeli M, Gasparetto P, Menotti RM, Pasuto A, Silvano S, Uniti CS et al (1996) A visco-plastic

model for slope analysis applied to a mudslide in Cortina d’Ampezzo, Italy. Eng Geol

  29:233–240

Angeli G, Buma J, Gasparetto P, Pasuto A (1998) A combined hillslope hydrology/stability

model for low-gradient clay slopes in the Italian Dolomites. Eng Geol 49:1–13

Aras A (2007) Karakteristik aliran sungai Serayu di bendung gerak Serayu Kabupaten Banyumas,

Jawa Tengah. Undergraduate thesis. Civil and environmental engineering department. Gadjah Mada University. Yogyakarta.

Condon WH, Pardyanto L, Ketner KB, Amin TC, Gafoer S, Samodra H (1996) Geological map

of the Banjanegara and Pekalongan Sheet, Jawa. Geological Research and Development

  Centre, Bandung

Corominas J, Moya J, Ledesma A, Lloret A, Gili JA (2005) Prediction of ground displacements

and velocities from groundwater level changes at the Vallcebre landslide (Eastern Pyrenees, Spain). Landslides, (Oct 2004), pp 83–96

  

Fathani TF, Karnawati D, Sassa K, Fukuoka H, Honda K (2008) Landslide monitoring, prediction

and early warning in Banjarnegara, Indonesia. Star, 4–6 Sept 2008

Hong Y, Hiura H, Shino K, Sassa K, Suemine A, Fukuoka H et al (2005) The influence of intense

rainfall on the activity of large-scale crystalline schist landslides in Shikoku Island Japan.

  Landslides 2:97–105

Karnawati D, Fathani TF, Sudarno I, Andayani B (2008) Development of community-based

landslide early warning system in Indonesia. In: Proceeding of the first world landslide forum, Tokyo, 18–21 Sept 2008, pp 305–308

  

Okada K, Makihara Y, Shimpo A, Nagata K, Kunitsugu M, Saito K (2001) Soil water index.

  Tenki 47:36–41

  

A Coupled Hydrology/Slope Kinematics Model 467

Ranalli M, Gottardi G, Medina-cetina Z, Nadim F (2009) Uncertainty quantification in the

calibration of a dynamic viscoplastic model of slow slope movements. Landslides 7(1):31–41

Rokhmana CA (2007) The low cost monitoring system for landslide and volcano with digital

photogrammetry. In: Proceedings of Joint Convention HAGI-IAGI-IATMI, Bali

Takeuchi A (1981) Method of investigating groundwater-vein-streams by measuring one-meter-

depth in landslide areas (II). J Jpn Assoc Groundw Hydrol 23(1):1–27

Van Asch T, Van Genuchten P (1990) A comparison between theoretical and measured creep

profiles of landslides. Geomorphology 3:45–55