Acta Geodyn. Geomater.Vol.2, No.1 (137), 25-28, 2005

GEODETIC DEFORMATION MONITORING IN ZAFARRAYA FAULT AND SIERRA TEJEDA ANTIFORM (SPAIN):

STATUS REPORT

Antonio J. GIL 1)*, Jesús GALINDO-ZALDÍVAR 2), María Jesús BORQUE 1), Carlos Marín-LECHADO 3), Patricia RUANO 2) and Carlos Sanz de GALDEANO 4)

1) Dept. Ing. Cartográfica, Geodésica y Fotogrametría. E.P.S. Universidad de Jaén, Spain, Campus Las Lagunillas, 23071 Jaén, Spain

2) Dept. Geodinámica, Universidad de Granada. 18071 Granada, Spain 3) Instituto Geológico y Minero de España, Spain 4) Instituto Andaluz de Ciencias de la Tierra CSIC- Universidad de Granada. 18071 Granada, Spain *Corresponding author‘s e-mail: ajgil@ujaen.es (Received July 2004, accepted November 2004) ABSTRACT The NW-SE to NNW-SSE convergence between the African and Eurasian plates in the western Mediterranean has developedthe recent relief of the Betic-Rif Cordilleras. The central part of the Internal Zones of the Betic Cordilleras is deformed bylarge open folds and faults, mainly with normal character. The Zafarraya fault, located to the N of the Sierra Tejeda antiform,was active during the 1884 Andalusia earthquake of 6.7 estimated magnitude. In the framework of an interdisciplinaryresearch project a non-permanent GPS-network has been designed at Zafarraya Fault and Sierra Tejeda antiform to monitordeformations related to these active tectonic structures. KEYWORDS: Active Tectonics, GPS Network, Zafarraya Fault, Sierra Tejeda Antiform, Betic Cordillera

sediments. This fault shows evidences of activity during the 1884 Christmas Andalusia earthquake with an estimated magnitude near 6.7 (Muñoz and Udías, 1980).

2. GPS NETWORK FOR DEFORMATION

MONITORING In order to quantify the deformation that is

currently occurring in the Zafarraya Fault and Sierra Tejeda antiform, the first non-permanent GPS network of this area has been designed (Fig. 1). The best option to determine very small movements using geodetic measurements is a network of permanent GPS stations. However, when this is not possible due to financial constraints, the network of closely-spaced pillars and its periodic observation by GPS campaigns spaced over time becomes the best option (Betti et al., 1999).

The network design and the individual site selection take into account geological, logistical and observational aspects. As a very long-term project is considered the monumentation should guarantee the reoccupation after a time span. This network is made up of sixteen reinforced concrete pillars anchored to rock (Figs. 1 and 2). Moreover, to assure that the antennas will be placed exactly at the same position in different reoccupations, the pillars have an embedded forced centering system (Fig. 2).

1. INTRODUCTION The Betic-Rif Cordilleras are build-up in the

western Mediterranean by the deformations related tothe Eurasian-African plate boundary. The recent NW-SE convergence (De Mets et al., 1990) produces thesimultaneous development of large folds and faults inthe central part of the Internal Zones of the BeticCordilleras, which continues active up to present(Galindo-Zaldívar et al., 2003). The uplift of themountain ranges is mainly related to the developmentof folds in this regional compressive setting.However, the most abundant faults recognised onsurface in the central part of the Cordillera shownormal slips, sometimes with dextral or sinistralcomponents.

Sierra Tejeda (Fig. 1) constitutes a relief relatedto the development of a complex shaped by recentantiform that is located near the coastline, overprintedon previous folded Alpujarride rocks belonging to theInternal Zones of the Betic Cordilleras. Along thenorthern limb of Sierra Tejeda is located the contactbetween the Internal and External zones.Geomorphological evidences, like the incision of thefluvial network or the cliff development along thecoastline, indicate its recent uplift. At the northernlimb of Sierra Tejeda is located the Zafarraya fault, ahigh-angle northwards dipping normal fault, whichproduces the development of a half graben in itshanging wall filled by Miocene-to-Quaternary

A. J. Gil et al.

26

Fig. 1 GPS network in Zafarraya Fault and Sierra Tejeda antiform

Fig. 2 a) Reinforced concrete pillar anchored to rock in site 800 b) Detailed view of the force centering device

GEODETIC DEFORMATION MONITORING IN ZAFARRAYA FAULT AND SIERRA TEJEDA …

27

3. PROCEDURE TO TEST THE COORDINATE DIFFERENCES The definition of a test to compare two GPS

networks is based on the assumption that the observations, Cartesian components of the GPS baselines, the coordinate estimates and their differences are normally distributed. The significance of the coordinate differences is tested by the well-known statistic:

1

20

ˆ ˆˆ

t

sp

x Q xF

mδδδ δσ

−

= (1)

where Qδδ is the cofactor matrix of the coordinate dif-ferences x̂δ , m the number of tested parameters (coordinate differences of points), 2

0σ̂ the variance of

unit weight obtained from 1r , 2r , 201σ̂ , 2

02σ̂ which are, respectively, the redundancies and the variances of unit weight of the two adjustments, α),(, 21 rrmsp FF +≈

(Fisher distribution) if the null hypothesis Ho: 21 xx ˆˆ = is true. If:

trrmsp FFF =≤ + α),(, 21 (2)

the null hypothesis Ho is accepted, and the coordinatesdifferences show no significant change. On the contrary if,

trrmsp FFF => + α),(, 21 (3)

the null hypothesis is rejected and it is possible to conclude that the coordinates have changed significantly.

To identify a point whose coordinates have changed it is necessary to test each point separately in turn (Crespi, 1996). To establish the final datum, i.e. to find the stable points, it is necessary to use an iterative procedure. First, each point is tested to define a starting datum including only the point i for which the statistics (1) is minimum. Next, the coordinate differences and their covariance matrix in the current datum are computed. Then, new test on each point to choose other point j to be included in the datum is carried out for which hypothesis Ho is accepted:

α),(,)( 21 rrmjsp KFF +≤ where mj is the number of point j

coordinates tested. At the first iteration all points except the fixed point i are tested. The procedure continues starting again in the second step and stops when the datum cannot be enlarged.

It is important to stress that a complete spatial rotation between the two sets of coordinates taking into account their full covariance matrices should beestimated to reduce the coordinates to the same reference frame before testing the statistical significance of the coordinate differences.

The GPS sites (Fig. 1) are located in a local anda regional network in order to study the local motionalong the Zafarraya fault and the regionaldevelopment of the Sierra Tejeda antiform. In thelocal network the sites 811, 812 and 816 are locatedon the hanging wall of the fault. Most of them arebuild up on Jurassic limestones of the External Zones.The sites 810, 813, 814 and 815, located in thefootwall, are build-up mainly on limestones andmarbles. Although these sites are located across thecontact between External and Internal Zones, thiscontact is inactive at present, and the southern part ofthe network may constitute a reference for the activityof the most recent Zafarraya Fault. The site 800 is inthe northern boundary of the regional network onJurassic limestones of the External Zones. South-wards, the network crosses the Zafarraya Fault and thecontact between External and Internal Zones andreaches the uppermost part of Sierra Tejeda in the site850, located on Triassic Alpujarride marbles. Thesouthern part of the network is located along thesouthern limb of Sierra Tejeda, deformed by NW-SE oriented normal faults, and reaches the coastline withsites 890 and 880, build up respectively onAlpujarride metapelites and marbles. The regionalnetwork also covers the WNW periclinal end of theSierra Tejeda antiform (Fig. 1). All sites meet thefollowing requirements: no obstruction above 10degrees; no high power lines nearby; easily accessible,etc.

The first survey will be done in September 2004.The second survey will be done two years after, inSeptember 2006. The GPS constellation will betracked during five-day campaigns with eight-hour sessions over baseline lengths ranging from 2 km to20 km. The equipment will be 6 Leica GX1230 dualfrequency carrier phase GPS receivers.

The network will be tied to the EUREF sitesAlmería, San Fernando and Lagos computing the coordinates of point 860. GPS data processing will beperformed by using Bernese software (Rothacher etal., 1996) computing single sessions in multibaselinemode. The first step (preprocessing) related toreceivers clocks calibration, will be performed by code pseudoranges, and detection and repair of cycleslips and removal of outliers, will be carried outsimultaneously for L1 and L2 data. The final solutionfor each session will be obtained using the iono-free observable with precise ephemeris and antenna phasecentre variation files. The coordinates of the stationsand relative integer ambiguities will be estimated inboth campaigns. From these results we will use anintermediate program to produce GPS baselines withtheir covariance matrixes. Using the NETGPSprogram that performs the adjustment of GPSbaselines accounting for their full covariance matrixes, the minimal-constrained network adjustmentwill be done.

A. J. Gil et al.

28

REFERENCES Betti, B., Biagi, L., Crespi, M. and Riguzzi, F.: 1999,

GPS Sensitivity Analysis Applied to non-permanent Deformation Control Networks. Journal of Geodesy, 73 (3), 158-167.

Crespi, M.: 1996, Software Package for the Adjustment and the Analysis of GPS Control Networks. Reports on Survey and Geodesy in memoria of Proff. A. Gubellini and G. Folloni. (Ed. M. Unguendoli), Edizione Nautilus, Bologna, 237-264.

De Mets, C., Gordon, R.G., Argus, D.F. and Stein, S.: 1990, Current plate motions. Geophysical Journal International, 101, 425-478.

Galindo-Zaldívar, J., Gil, A.J., Borque, M.J., González-Lodeiro, F., Jabaloy, A., Marín-Lechado, C., Ruano, P. and Sanz de Galdeano, C.: 2003, Active faulting in the internal zones of the central Betic Cordilleras (SE, Spain), Journal of Geodynamics 36, 239–250.

Muñoz, M. and Udías, A.: 1980, Estudio de los Parámetros y Serie de Réplicas del Terremoto de Andalucía del 25 de Diciembre de 1884 y de la Sismicidad de la Región Málaga-Granada. In: El Terremoto de Andalucía de 1884, Instituto Geográfico Nacional (ed.), Madrid, 95-139.

Rothacher, M., Mervant, L., Beutler, G., Brockmann, E., Fankhauser, S., Gurtner, W., Johnson, J., Schaer, S., Springer, T. and Weber, R.: 1996, BERNESE GPS software version 4.0. Astronomical Institute University of Berne.

The DENETGPS program (Crespi, 1996) will beused to evaluate the significance of the coordinatesdifferences without assuming any initial hypothesis onthe point’s behaviour or any geophysical constraint.This assumption corresponds to fix the centroid of thewhole network.

4. CONCLUSIONS

A GPS monitoring network has been establishedin Sierra Tejeda antiform and Zafarraya fault, and thefirst two campaigns are planned to take place inSeptember 2004 and 2006. A clear statisticalprocedure will be used to evaluate the significance ofthe coordinate differences without assuming anyinitial hypothesis on the point’s behaviour or anygeophysical constraint. The movements are probablyless than 1 mm per year and may have also acoseismic character. The reoccupation campaignwhich will be carried out two years after the firstcampaign will allow confirming that the rates aresmall. In order to detect such small rates, one willhave to measure over a longer time span, probably adecade.