1. Paleoseismology (Russia: S. Arjannikov; France: J-F. Ritz)
The goal of our study was twofold: To determine what is the kinematics of the active faults in the Tunka-Sayan area and to quantify the slip rates and the mean return period of earthquakes along the faults. Our approach has been the tectonic geomorphology and the paleoseismology (trenching). The first field studies in 1999 and 2000 had led to select two sites along which we had observed evidences of recent tectonic inversion (i.e. from transtension to transpression): The E-W Mondy valley which connects the Tunka basin to the Hovsgol rift system (Larroque et al., 2001) and the North Tunka fault itself, at one site called Bielly Kamen.
Several questions remained to better characterize the active tectonics at these sites: What was the precise geometry (length, segmentation, kinematics) of the Mondy fault zone including the Ikhe-Ukghun secondary fault located just to the North; and how many events were responsible of the Bielly Kanen fault scarp along the North Tunka fault ?
These questions have been the subject of Sergey Arjannikov’s work: during summer 2001 during a field study in collaboration with his colleagues V. Sankov and A. Arjannikova (also from the Institute of Earth’s Crust in Irkutsk) and this Autumn 2003 in Montpellier where S. Arjannikov stayed for 4 weeks.
The field data collected last summer were discussed and interpreted in Montpellier. They show that the Mondy fault is a major left strike-slip fault with a small reverse component, extending from the eastern edge of the Tunka basin to the Hövsgöl and dipping steeply to the South rift (Publication 3). In Bielly Kamen the re-occupation of the trench shows that the fault scarp correspond to only one big event (with therefore a preliminary estimated magnitude M ~ 8). The complementary sampling will allow to bracket this major event which age is estimated at 6.7 ka after our preliminary radiocarbon dates.
His stay in Montpellier allowed S. Arjannikov to update his knowledge in paleoseismology and seismotectonics through fruitful discussions and numerous visits to our Department library. We also took the opportunity of his visit to analyse one of the major question which remains unsolved in the Northwestern Sayan region: When did the Tertiary compression started up there ? This area, where S. Arjannikov collected already quite a few field data, that we started to synthesize in Montpellier, should be the subject of a future field investigation.

2. GPS (continuous) geodesy (Russia: A. Lukhnev; France: E. Calais)
Our objective was to provide technology and knowledge-how transfer to the Institute of the Earth’s Crust (IEC) so that they could become a GPS data processing center for continuously operating GPS stations in Asia for crustal deformation studies. We hosted Andrei Lukhnev on January 2002 and provided training in the use of the GAMIT-GLOBK GPS analysis software for the processing of continuous GPS networks. In summer 2002, we provided the IEC with a suite of Unix scripts to automate the processing of continuous networks. Since the IEC was already running the GAMIT-GLOBK software for processing campaign GPS measurements, the upgrade to processing data from permanent stations was rather straightforward. We also provided information and references on the computation of position time series for continuous GPS stations and their analysis in terms of data noise and geophysical signals. We provided software and training for the interpretation GPS velocities in terms of strain rates. We have maintained continuous contact by email since the beginning of the partnership, with technical and scientific exchanges. We have followed the progress made by A. Lukhnev in the use of these various piece of software and are confident that he is now proficient in their use.
Results of permanent GPS data are used in Publications 1 and 4. They indicate that crustal motions in Mongolia Baikal are significantly faster than the predictions of most deformation models of Asia. They also provide a key control on modelling post-seismic deformation caused by visco-elastic relaxation in the lower crust following the Bolnay-Tsetserleg earthquake sequence (Mw=8.4 and Mw=7.9, July 1905) as shown in Publication 1. Furthermore, we have quantified east- to south-eastward motion and left-lateral shear for central and eastern Mongolia which is accommodated by left-lateral slip on the E-W trending Tunka, Bolnay, and Gobi Altay faults (2 ? 1.2 mm/yr, 2.6 ? 1.0 mm/yr, and 1.2 mm/yr, respectively) and by about 4 mm/yr of extension across the Baikal rift zone (Publication 4). These values are key parameters for seismic risk assessments on these major faults.

Remarks, if any: