MANTLE FLOW IN THE CARPATHIAN BEND ZONE? INTEGRATION OF GPS AND GEOPHYSICAL INVESTIGATIONS
The unusual strong and persistent intermediate depth seismicity and surface deformation at and around the most tightly curved portion of the Carpathian arc in Central – Eastern Europe, Vrancea Bend Zone (approximate central co-ordinates 45.50N, 25.50E), has been the main target of numerous recent studies. They culminated in a persistent debate on the nature of the processes involved. The possible tectonic scenarios explaining the Vrancea seismicity and surrounding structures invokes (a) subduction of Tethys oceanic lithosphere, a remnant of which is presumed to be just detaching from unsubductible continental lithosphere of the East European and Moesian platforms; (b) oceanic lithosphere subduction ending some time in the late Miocene, and that since then a portion of East European or Moesian platform continental mantle lithosphere has been delaminating along a horizontal mid-lithospheric interface and dripping down into the upper mantle.
In order to add a robust, kinematic component to the ongoing studies, the University of Bucharest in close cooperation with The National Institute for Earth Physics and the universities from Delft, Utrecht and Karlsruhe have shaped and implemented a two decade-long program to use the modern satellite geodesy techniques in order to determine de directions and magnitude of velocity vectors for the area nearby Vrancea. We have carried out 19 GPS campaign and run a network of eight permanent stations used as reference frame in order to get a robust determination of velocity values which are so small that could be considered close to technical boundaries.
The very small deformations imposed a particular data processing. The daily solutions in terms of spatial coordinates are determined for each campaign in respect to an average solution of the independent network. This is done in an iterative mode, by decreasing the weight of the reference stations outside the network, using pre-defined criteria for each component, as a function of the statistical analysis of the GPS data set. The sigma values for the solutions of every location from every campaign have been calculated as a function of the variation interval for the daily solution in every location. In addition, secondary processing effects are being computed and their weight in the final solutions has been diminished. The campaign solutions have been afterwards used to estimate the velocity vectors for every component in each site. The average tendency has been determined for every location, as well as the campaign individual solutions using an offset higher than the pre-defined criterion. This iteration continued until getting the velocity vectors by components. In such a way the authors got a solution within the framework of the ITRF-2000 reference net. For the further geophysical analysis, the Eurasian rotation pole has been computed (56.310º N, 101.732ºW and 0.2569º/Ma) which has been elliminated from the solution. In this way, the resulting GPS velocity vectors are refered to the stable Eurasian continent.
The present outcome show that the ESE part of the Romanian territory gets away in respect to the stable Eurasia, velocity values reaching no more than 2.0 – 2.5 mm/y. In respect to the stable Carpathian vorland the movements are very slow and have to be confirmed by further campaigns to be completed on an annual base. Three sites east of the Carpathian arc show a bit higher velocities and the displacements are decreasing westwards. These differential displacements cannot be triggered by superficial landslides as they are located into perfect planes. However, the displacements are determined in front of the Carpathian arc, close to the intermediate depth seismogenic zone. The vectors could suggest a rapid displacement of the the lithospheric slab in respect to the Moesian Platform. The sites placed at the internal part of the Carpathain arc and within the Transylvanian Basin do not show significant displacements in respect with the east-southeast sites reference net but suggest a small counterclockwise rotation of this area. If we take into account the southeastwards position of the velocity vectors from the central-eastern part of the Moesian Platform, as well as the northwestwards position of the velocity vectors at stations in North Dobrogea and east of the prolongation of Peceneaga-Camena crustal fault system, we coul estimate a pretty robust astenospheric rotation flow around the Vrancea lithospheric slab.
Finally, taking into account the results from the high seismic tomography, seismic attenuation and satellite geodesy studies, we shaped the main kinematics of the Carpathian Bend Zone, suggesting an anti-clockwise displacement of the surface as being a result of deep, mantle flow, accompanied by an advancement of the melting processes at the astenosphere-high velocity body boundaries.
Acknowledgments. This cooperative works has been carried out during more than 15 years with significant support from ISES, Romanian Ministry of Education and Research and National Science Foundation (USA), SFB 461 (Germany), and with a coherent input by a prof. B.A.C. Ambrosius (TU Delft), Dr. Ray Russo (University of Florida) and tens of students from various universities in Europe, mainly from the University of Bucharest, TU Delft, Free University Amsterdam, University of Karlsruhe and others.