千葉県北部における地盤沈下検出のためのSAR干渉応用

Zhaogui ZHANG, Isao SATO and Minoru URAI(Geological Survey of Japan)
張 昭貴，佐藤 功，浦井 稔(地質調査所)
E-mail: Zhaogui ZHANG, zzg@gsj.go.jp

要旨：地質構造運動や人間活動で引き起こされる地盤沈下は，干渉SARにより検出可能で� る．ERS1およびJERS1のSARデータにより，千葉北部の地盤沈下を検出した．1992年―1998年の期間に取得されたERS1やJERS1のSARデータでの位相縞を干渉SARで求めた結果，インタフェログラムが地下水利用などによって生じていると考えられるこの地域での地盤沈下地域を反映していると推定された．干渉SARは今後の地質環境の保全に役立つものと考えられる．

Key words: SAR, interferometry, subsidence, Chiba prefecture

1. INTRODUCATION
The study area is located in northern Chiba prefecture (see P18fig1.jpeg) which covers 140^o -141^o E and 35.5^o-36^o N and the total area is about 5000 km2, where vegetation covers almost ground surface and agriculture is also well developed. Recent environmental investigation shows there are week subsidence near Tokyo bay area but no obvious and direct evidence to prove it. The launch of SAR sensor provided effective tool to detect and monitor the tiny deformation. The target of this study is to detect subsidence with interferometry technique.

fig1.  Location of study Area in Chiba Prefecture

The ideal to monitoring surface changes arose in early 1950. The geophysical application of satellite radar interferometry to detect changes in the earth surface has been exploded in early 1990s. John G. Curlander (1991)1 first outlined the progress of SAR systems and signal processing especially for Seasat-A SAR system and Didier Massonnet (1995 and 1998 )2, 3 systematically integrated the application of radar interferometry in changes of earth's surface. A successful application includes SAR data pairs, processing technique and an event.

2. DATA SELECTION AND PROCESSING

Land deformations can be classified by rate and duration caused by mankind activity or structure movement. Earthquakes, for example, displace the surface with high rate (~1m/s) but in short duration. At the other extreme, the post-seismic and normal structural activity are slow in rate (~1 mm /yr) but in long duration (101-103). To detect subsidence in an area, the most important research work is to define the subsidence rate and duration in recent years. Based on the materials of subsidence monitoring activities in the study area, two pair SAR (one is Ees1 and others are Jers1) data around 1994 are chosen and usual time span is 3 to 3.5 years(see table 1). Data processing in this study is carried out in EarthView Advanced Precision Processor (APP) and InSAR (version 1.0) software with differential mode based on DEM data (50m resolution). During the schedule of interferometric processing, co-registration was done by manual control point because the time gap of Jers1 pair usually are larger than three years.

Table 1.  SAR data  pair used in the study

3. INTERFEROMETRY FEATURE IN RESIDENCE AREA

Fig2. and fig3 show interferogram of Ers1 SAR in the north part of Chiba prefecture.

Fig.2.  ERS1 interferogram of Chiba prefecture acquired on Nov. 8,1992 and March 5,1996 with poor coherence.
 
 

Based on the above interferogram imageries, features in phase change area a, b, c, d, e, F and G are illustrated in table 2. It should be pointed out that area E and F form large phase change areas which are larger than 600 Km2. There are various kinds of possibilities to form phase changes, i.e., surface deformation. The first possibility is land development. The second is subsidence relating ground water extraction for drinking water supply, agricultural use, and industrial use. The third is gas production along the Kuju-kuri coastline where gas fields are distributed.
 
 

Table 2. The characteristics of phase changes in A, B, C, D, E and F area

location	Area (Km2)	Fringes (cycle) in Ers1	Fringes (cycle) in Jers1
A	27.34	1	Out of data coverage
B	12.31	1	Out of data coverage
C	15.84	1	Out of data coverage
D	240	<1	<1/2
E	24.96	1-2	<1
F	32.1	1-2	<1
G		not clear	<1

It is difficult to distinguish these factors from this study. Based on table 2 phase changed area E and F are caused by subsidence with a movement about 3-11.5cm in about 3.5 years.

4. CONCLUSION AND DISCUSSION

The SAR processing and analysis shows there are clear subsidence in northern Chiba area. It should pay special attention to these areas and also need to validate. The residual fringe appeared in D, if confirmed, seems to be caused by other mechanism, such as plate movement, which is belongs to another projects.

5. ACKNOWLEDGEMENT

The author would like to thank AIST and GSJ officers and scientists who help me a great and support me to work and study in Geologic Remote Sensing Section, Environmental Geology Department in Geological Survey of Japan and also to thank the supporters of DEM data (made by GSI) and SAR data (received and processed by NASDA) for research use.

6. REFERENCE

1. John G. Curlander, Synthetic Aperture Radar-Systems and Signal Processing, John Wiley & Sons, Inc., 1990.
2. Didier Massonnet, Deflation of Mount Etna monitored by spaceborne radar interferometry.
3. Didier Massonnet and Kurt l. Feigl, Radar interferometry and its application to changes in the earth 's surface, Review of Geophysics, 36, 4/Nov. 1998, 441-500, 1998.