Land Subsidence at the Kujukuri Plain in Chiba Prefecture, Japan: Evaluation and Monitoring Environmental Impacts

Home , Chiba Prefecture, Mobara, Shirako

Land Subsidence, Associated Hazards and the Role of Natural Resources Development 293 (Proceedings of EISOLS 2010, Querétaro, Mexico, 17–22 October 2010). IAHS Publ. 339, 2010.

Land subsidence at the Kujukuri Plain in Chiba Prefecture, Japan: evaluation and monitoring environmental impacts

H. OBANAWA1, T. TOKUNAGA1, S. ROKUGAWA2, T. DEGUCHI3 & T. NAKAMURA2 1 Graduate School of Frontier Sciences, The University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa-shi, Chiba, 277-8563, Japan [email protected] 2 Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8656, Japan 3 Nittetsu Mining Consultants Co., Ltd., 4-2-3 Shiba, Minato-ku, Tokyo, 108-0014, Japan

Abstract The impacts of land subsidence on surface environmental changes were analysed for the Kujukuri Plain, Japan, where subsurface brine has been abstracted for more than 50 years. Reconstruction of past landforms using airborne laser scanning and existing levelling data showed that the geomorphological setting of the whole region, such as beach ridges and backmarshes, has been retained during the recent 40 years. Apparent and simple relationships between shoreline retreat and land subsidence were not recognized, and the spatio-temporal distribution of shoreline change seems to be mainly controlled by coastal constructions. In addition, a method to measure long-term deformation by combining InSAR and time series analysis was proposed to improve our ability for monitoring land subsidence. The local uplifts and subsidence detected by the proposed method were quite consistent with those obtained from levelling data, suggesting that our method was applicable to the detailed monitoring. By combining and integrating the approaches presented in this paper with numerical modelling of deformation/fluid flow processes, scientifically sound recommendations for the sustainable development of the subsurface resources can be proposed for society. Key words land subsidence; environmental impact; monitoring; InSAR; GIS; Kujukuri Plain, Japan

INTRODUCTION Land subsidence causes various types of damage to civil society and it may result in the permanent deterioration of environmental conditions due to the irreducible nature of the subsidence phenomena in general (Deming, 2002). Because of the industrial withdrawals of subsurface fluid resources, such as dissolved methane and groundwater, land subsidence has occurred at many places in Japan (Environmental Management Bureau, Ministry of the Environment, 2009). The Kujukuri Plain, Chiba Prefecture (Fig. 1), is one of the places where heavy subsidence, with a maximum value of accumulated subsidence of 100.7 cm, was observed during the period from 1969 to 2007 (Environmental and Community Affairs Department, Chiba Prefecture, 1970–2007). This coastal plain is composed of alternating beach ridges and backmarshes (Moriwaki, 1979); (Fig. 1). Most of the area is situated below 10 m elevation, therefore it is of concern that continuous land subsidence by abstraction of subsurface brine and possible future sea-level rise due to global warming may increase the risks of flood disaster and coastal erosion. However, our understanding of the surface environmental changes caused by subsidence are not yet sufficient to predict the extent of the environmental impacts and to propose efficient countermeasures against these influences. This paper presents our activities to tackle the problems through two different approaches, i.e. (1) to better understand the surface environmental changes caused by land subsidence, and (2) to further improve the Interferometric Synthetic Aperture Radar (InSAR) method to achieve a high resolution monitoring technique for surface movement. We also present the importance of integrating these approaches to achieve the sustainable development of subsurface resources.

294 H. Obanawa et al.

138° 140° 38°

36°

Katakai Fishing Port

Makame River

Nabaki River Plateau Beach ridge Ichinomiya River Sand dune 0 5km Backmarsh Taito Fishing Port Taito-misaki Fig. 1 Geomorphological map of the Kujukuri Plain (modified after Moriwaki, 1979). Square marked by dashed line indicates the area of Fig. 2.

ANALYSING CHANGES OF THE SURFACE ENVIRONMENT Reconstructing past landforms We conducted an airborne laser scanning survey in December 2008 to obtain high accuracy and high resolution digital elevation model (DEM) (Fig. 2). The study area includes Mobara City, Shirako Town and Chosei Village on the southern Kujukuri Plain. The DEM has an area of 224 km2 (14 km × 16 km), 1-m mesh resolution, and 4.0-cm vertical error, which is much higher in accuracy than the existing topographic map and the standard specification for airborne laser scanning, i.e. 25 cm (Ministry of Land, Infrastructure, Transport and Tourism, 2008). We estimated landform changes by integrating the present DEM-based topographic data and the existing levelling data. We used the “Precise levelling survey results in Chiba Prefecture” compiled by Chiba Prefectural Government (Environmental and Community Affairs Department, Chiba Prefecture, 1970–2007) to reconstruct the past ground level. The levelling survey has been carried out every year since 1969. The maximum value of accumulated subsidence was 100.7 cm in the northeastern part of the Mobara City and the minimum value was 39.2 cm in the southern part of Chosei Village during the period from 1969 to 2007. The landform changes were obtained by subtracting the ground-level changes from the present DEM in the study area.

Expansion of lowland Figure 3(a) shows the expansion of lowland as a function of time. The area below 1 m was 2.4% of the whole study area in 1969, while it has expanded to 5.5% in 2009. Areas below 2 to 6 m have increased more rapidly. For example, the area below 6 m increased by 5.6% during the period 1969– 2009. Lowland has expanded along the Nabaki River and partially in the swale behind the beach ridge along the shoreline (Fig. 3(b)). However, overall the spatial pattern of the geomorphological features such as beach ridges and backmarshes has been retained, i.e. regional difference of land subsidence has not caused significant change of the entire landforms during the last 40 years. Land subsidence at the Kujukuri Plain in Chiba Prefecture, Japan 295

Fig. 2 One metre mesh digital elevation model obtained from the airborne laser scanning survey.

(a) (b) Fig. 3 Temporal change of landform in the past 40 years. (a) Altitude-area relations; and (b) an example of the landform change at the lower reach of the Nabaki River.

296 H. Obanawa et al.

Shoreline change The existing data related to shoreline change were summarized, and we compared the spatio- temporal changes of the shorelines with those of ground levels (Fig. 4). In Fig. 4(a), the sites and the timings of the installation of coastal architectures such as training dikes, jetties and break- waters were also shown. The apparent and simple relationships between the shoreline retreat and the land subsidence (Fig. 4(b)) were not observed, i.e. the areas with a large amount of accumulated shoreline retreat do not correspond to those with a large amount of land subsidence. Spatio-temporal changes of the shoreline seem to be controlled more by artificial coastal structures. For example, the area of shoreline retreat has expanded after the completion of the construction of the breakwater in front of the sea cliff at Taito-misaki (see Fig. 1 for the location) at 1985, which is considered to be a major source of beach sand at the study area. In addition, the saw-like pattern of alternating high and low erosion areas are observed in between the Ichinomiya River and the Taito Fishing Port. These patterns are interpreted to be caused by the construction of headlands in the area.

Amount of accumulated Coastal strucures shoreline change (m) Training dike 120～ 60～120 Jetty or headland 1～ 60 -1～ 1 -60～ -1 Breakwater A: Shoreline change (Reference year: 1969) -120～ -60 ～-120 North ← The distance measured along the coast from Iioka Fishing Port (km) → South

32 3334 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55Construction start Year 1970 Wave-dissipating dikes at Taito-misaki 1980

Protected whole area 1990

2000 No No data data

Katakai Fishing Port Makame River Hori River Nabaki River Ichinomiya River Taito Fishing Port Current direction of littral drift Year 1970 Amount of accumulated B: Ground level change land subsidence (mm) 1980 -50～ 0 -100～ -50 -150～-100 1990 -200～-150 -250～-200 2000 -300～-250 -350～-300 -400～-350 -450～-400 Hori River NabakiRiver Ichinomiya River -500～-450 -550～-500 32 3334 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 -600～-550 North ← The distance measured along the coast from Iioka Fishing Port (km) → South -650～-600 Fig. 4 X-t diagrams showing spatio-temporal changes of shoreline and ground level. The vertical axis indicates time and the horizontal axis distance along the coast. The gradation from white to black indicates the amount of accumulated shoreline change in the upper figure (A) and ground level change in the lower figure (B). The year 1969 was chosen as a reference to construct the diagrams. Solid and dashed lines in the upper figure (A) show the coastal structures.

IMPROVEMENT OF INSAR TECHNIQUE TO MONITOR GROUND DISPLACEMENT Interferometric Synthetic Aperture Radar (InSAR) is becoming established as a method for monitoring ground displacement that can observe subtle surface movement over a wide area at high resolution (e.g. Ferretti et al., 2001; Berardino et al., 2002). We recently developed a method to measure long-term deformation by combining InSAR and time series analysis, aiming at estab- lishing the practical and flexible measurement technique (Deguchi et al., 2009). In our measuring method, the value of land deformation is set as an unknown parameter. Then, by applying a smoothness-constrained inversion algorithm, the optimal solution to the amount of land deformation was obtained. We applied our new method at the study area with ENVISAT/ASAR data, and verified its accuracy by comparing our results with levelling data (Fig. 5). It was shown that the local uplifts and subsidence detected by the proposed method were quite consistent with those obtained from levelling data both in spatial distribution and the amount of vertical movement. Land subsidence at the Kujukuri Plain in Chiba Prefecture, Japan 297

InSAR Levelling data

Fig. 5 Comparison among InSAR (left), levelling survey (centre), and the temporal changes of surface deformation obtained from our InSAR analysis (dots) and levelling survey results (triangles) (right).

FUTURE PERSPECTIVES To evaluate and monitor the environmental impacts by subsurface resources abstraction, integration of the detailed analysis of surface topographic change with GIS platform and the detailed monitoring by InSAR technique is considered to be very effective. Combination of our approaches with numerical modelling of land subsidence by resources abstraction will provide us with reliable and scientifically sound recommendations for the sustainable production scheme, and it will help to build the possible consensus on the sustainable development (Fig. 6). The approach presented here is considered to be universal, and hence it will be used for the areas where similar subsurface development has been conducted and/or planned.

Modeling of subsurface fluid movement and Formulation of development and management plan ground deformation

Planning Prediction of surface environmental change

Prediction of ground deformation Scientific understanding

Observation program Modification and evaluation of the model GIS platform for predicting and evaluating ground Large-scale monitoring of ground deformation (e.g. InSAR) deformation and surface environmental change

Digitalization

Modeling (abstraction)

Real world

Action Visualization Observational Decision-making result

Fig. 6 Conceptual diagram showing the importance of integration of monitoring and modelling for sustainable development of subsurface resources. 298 H. Obanawa et al.

Acknowledgements We would like to express our sincere gratitude to the Keiyo Natural Gas Association for supporting this study.

REFERENCES Berardino, P., Fornaro, G., Lanari, R. & Sansosti, E. (2002) A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms. IEEE TG & RS 40, 2375–2382. Deming, D. (2002) Introduction to Hydrogeology. McGraw-Hill, New York, USA. Deguchi, T., Rokugawa, S. & Matsushima, J. (2009) Long-term ground deformation measurement by time series analysis for SAR interferometry. J. Remote Sensing Soc. Japan 29, 418–428 (in Japanese with English abstract). Environmental and Community Affairs Department, Chiba Prefecture (1970–2007) Levelling results in Chiba Prefecture [reference date: February 1970–January 1st, 2007], (in Japanese). Environmental and Community Affairs Department, Chiba Prefecture (2007) State of land subsidence in Chiba Prefecture (in Japanese). Environmental Management Bureau, Ministry of the Environment (2009) Summary of subsidence area in Japan, the year 2008 (in Japanese). Ferretti, A., Prati, C. & Rocca, F. (2000) Nonlinear subsidence rate estimation using permanent scatterers in differential SAR interferometry. IEEE TG & RS 38, 2202–2212. Ministry of Land, Infrastructure, Transport and Tourism (2008) Survey Act, Operation Rules (in Japanese). Moriwaki, H. (1979) The landform evolution of the Kujukuri Coastal Plain, Central Japan. The Quaternary Research 18, 1–16 (in Japanese).