Deformation Observations at Merapi Volcano, Java island, Indonesia by Using GPS DataAris Sunantyo Gadjah Mada University, Faculty of Engineering, Geodetic Department, Jl. Grafika No-2, Yogyakarta, Indonesia Email:sunantyo@yahoo.com Carl Gerstenecker Darmstadt University of Technology, Physical of Geodesy, Petersen Strase 13, Darmstadt, Germany.
Abstract:
Merapi volcano is a high andesitic stratovolcano and one of 129 very active volcanoes in Indonesia. It is located close to Yogyakarta city, Java island with more than 500 000 inhabitants. Deformation observations at Merapi volcano has been conducted by tilting, Electronic Distance Measurement (EDM) and GPS method. How to analyze deformation at Merapi volcano using repetition of GPS data will be the objective of research. The research used repetition of GPS observations which was carried out between 1998 and 2002 at Merapi volcano. The network consists of five continuous stations established by the GeoForschungsZentrum Potsdam (GFZ) and the Volcanological Survey of Indonesia (VSI) as well of - so called – 18 campaign stations established by ourselves. Campaign and continuous stations are observed during 4 field campaigns. For the observations we used up to 11 Trimble GPS receivers 4000 SSE and SSi with geodetic L1/L2 antennas. Data analysis has been carried out with Bernese software 4.2 using precise orbit ephemeris. The coordinates are determined in the ITRF 2000 reference frame defined by the IGS stations DARW, COCO, NTUS and BAKO. To determine local deformations, we transform the results of each campaign to each other using the 7 parameter Molodensky transformation approach. Differences between the transformed coordinates are interpreted as elastic deformations. The statistical significance is tested by applying global F-test and local t-test statistics. In general the result of research has found no deformations at the slopes of Merapi and at distances bigger than 2 km from the crater rim. The stations at the crater rim are changing with time and give some hints about active local faults at Merapi summit. 1. Introduction Merapi volcano is a high andesitic stratovolcano and one of 129 very active volcanoes in Indonesia. It is located close to Yogyakarta city, Java island with more than 500 000 inhabitants. It has been selected as one of 15 high risk volcanoes within the International Decade of Natural Disaster Reduction program (IDNDR) of UNESCO. Deformation of volcanic edifices can give detailed information about the actual state of the volcano. Deformation observations at Merapi volcano has been conducted by tilting, Electronic Distance Measurement (EDM) and GPS method. Tilt meter and EDM are strongly influenced by local effects as temperature changes or changes of atmospheric refraction, with GPS method will not influenced by local effects so, it seem to be promising method. All the measurements has been carried out to observe deformations at the crater rim. Deformation of volcanic edifices can give detailed information about the actual state of the volcano and it is part of research in every volcano in Indonesia (Purbawinata, et al, 1997). Deformation observations at Merapi volcano, Indonesia have started in the 80thies of the last century (e.g. Young et al., 2000a, Young et al., 2000b, Voight et al., 2000;), it can be observed by many instrument such as using tiltmeters (Ratdomopurbo and Andreastuti, 2000) and Electronic Distance Meter(EDM) and theolite (Sunantyo, 1997). Both methods are strongly influenced by local effects as temperature changes or changes of atmospheric refraction. Deformation is one of characteristic activity at Merapi volcano (Ratdomopurbo and Andreastuti, 2000). GPS has been used as one of instrumentation standard for monitoring of volcanoes in Indonesia (Ratdomopurbo dan Sampurno, 2000). All the measurements were carried out to observe deformations at the crater rim. In 1994 also GPS-observations (Beauducel et al., 1999) have started. With two geodetic GPS-receivers the deformations of a profile along the north flank of Merapi and of points at the crater rim have been observed. 1996 the GeoForschungsZentrum Potsdam has installed within the Indonesian German research project “MERAPI” (Zschau et al, 1999) the permanent network “PN”, consisting of five permanent GPS-stations. 1997 the Institute of Physical Geodesy, Darmstadt University of Technology, established a deformation network - called in the following “CN” (campaign network) - consisting of 23 stations (including the permanent stations) around the summit of Merapi volcano. The network was observed 6 times between 1997 and 2002. In the following sections we will describe the results of deformation observations for the observation campaigns carried out between August 1998 and August 2002. 2. Data Analysis 2.1.Permanent GFZ-Network (PN) The permanent GFZ- network consists of 5 stations. The base station MVO YOGYA is located at Merapi Volcano Observatory, Yogyakarta. Three further stations (KLAK, KEND and SELO) have been established in 1996. 1998 the GPS-station GMRR was installed. Figure 1 represents the station locations, in table 1 the approximate geographical coordinates are given. The stations are identical to the multi-parameter stations, which have been established by GFZ in co-operation with the Volcanological Survey of Indonesia (VSI) (Zschau et al., 1999). At the stations Trimble 4000 SSE and SSi receivers with geodetic L1/L2 antennas with groundplates have been installed. Because of the small memory size of the receivers (< 500KB) the time period for observations was limited: every day between 6 and 12 o’clock the measurements were carried out. The data were sent via telemetry to MVO every day. More details of observation arrangements are given in table 2. 2.2 Campaign Network (CN) The campaign network “CN” was established to monitor 3 - dimensional deformations in distances up to 50 km from the summit of Merapi volcano. Additionally we have observed gravity changes at these stations (e.g. Setiawan, 2003, Tiede, 2005, Jentzsch et al., 2005, Tiampo et al, 2004, Tiede et al, 2005). Location and approximate coordinates are shown again in Fig. 1 and Table 1. Details of observation arrangement are given in Table 2. For observation campaigns between 1997 and 1999 only 3 GPS- receivers Trimble 4000SSE and SSi have been available. 2000 and 2002 11 receivers have been used at the same time. 2.3.Reference Network For the computation of station coordinates on the WGS 84 ellipsoid within the ITRF2000 we have connected selected stations to the IGS-stations BAKO, COCO, DARW and NTUS. The configuration of this reference network is shown in fig. 3 and 4. The coordinates of the reference stations are listed in Table 3. 2.4. Steps of data analysis. Data analysis was carried out by four steps are follow :
To determine statistically significant displacements we have applied the global F-Test according with v : residual of all observations i : epoch i m : number of epochs vi : residual vector of observations of epoch i ni : degree of freedom for epoch i N : degree of freedom of all observations The F-test shows, that at all epochs statistically significant displacements at the 95% probability level were observed. To localize the displacements we have carried the 7 parameter transformation according Molodensky, Bursa and Wolf. In principal this transformation determines three translations and three rotations and a scale factor to map the coordinates of epoch i (start system) to the coordinates of the reference epoch (target system). The transformation is described by equations (2) and (3). As reference epoch we have chosen the coordinates observed during the field campaign 2000.  ---------------(2) ---------------(3)The transformation parameters were determined by least square adjustment iteratively. In the first adjustment we assume that no displacements between start and target system are present. The coordinates of all stations are used, to determine the transformation parameters. We receive according to equation (4) a residual vector res can be computed :  ---------------(4)The components of the vector res are tested again using the local t-test according  ---------------(5)with si,u = standard deviations of resi,u, u = x or y or z. at the 95% significance level. If t exceeds the threshold, the least square adjustment of the transformation parameters is repeated without the coordinate component u. 3.Results and discussion All of results of the research will be shown from Fig. 1 till Fig.4.  Figure 1: Statistically significant horizontal displacements between 1998 and 2002  Figure 2: Statistically significant horizontal displacements around Merapi summit, observed between 1998 and 2002  Figure 3: Vertical displacements  Figure 4: Vertical displacements around Merapi summit Regarding the plots from Fig. 2 till Fig.5 show, that we have observed large deformations only at the summit stations of the network. The displacements are especially between 2000 and 2002 very large (<200 mm). All stations at the summit show a movement away from the center of the crater. Vertical displacements are also significant. However the behavior of the summit stations is not uniform. Displacements vanish with increasing distance d of the stations from the crater rim. At stations where d > 2 km no significant displacements (horizontal as well as vertical) are observed. Displacements at stations like BOYO or BUTU are caused by tectonic plate movements and not by volcanic activity of Merapi. 4.Conclusion.
    Figure 2. Postfit LC phase residual for all the observed satellites (10 satellites) recorded at all the stations in Miyakejima. Red vertical bar indicates the start of the eruption, which was 17:02 JST according to (JMA, 2000). Red line is the residual for SV5, while cyan for SV9, green for SV30, yellow for SV23, blue for SV29, black for SV21, magenta for SV26 and magenta plus line for other satellites (SV6,17,25).     Figure 3. Sky plot of the satellites and their postfit phase residuals. a) Sky plot for SV5, SV23, and SV21 above Miyakejima volcano from 17.00 JST until 19.00 JST. Solid diamond represents the satellite track in the period of 17:15-18:15 JST. Open rectangular is the location of the abundant ash, reported by VRC-ERI. b) Postfit phase LC residuals for SV5 and c) SV23 d) SV21.     Figure 4. Evolution of the refractive index during the eruption estimated by Kalman filter. a) 17:06-17:09 JST, b) 17:09-17:12 JST, c)17:12-17:15 JST, d) 17:15-17:18 JST, e) 17:18-17:21 JST, f) 17:21-17:24 JST, g) 17:24-17:27 JST, h) 17:27-17:30 JST. The color bar indicates the refractive index (cm/m). Unit for the space model is in kilometer.
Page 1 of 1
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||