The Research Station RAS and the International Research Center-Geodynamic Proving Ground have started the implementation of a 2-year program of a “pilot” project of electromagnetic monitoring by means of two broadband measuring stations Phoenix MTU-5D System 2000. The project is being implemented under the agreement signed between Phoenix Company (Canada) and the Research Station RAS at the end of 2002. The monitoring is carried out at the expense of the Research Station RAS using its internal resources. Phoenix has gratuitously delivered two kits of its equipment to the Research Station for the period of two years. Such mutually beneficial agreement became possible due to the high level and international prestige of magnetotelluric investigations carried out by the Research Station RAS for 20 years.

    The project is aimed at the creation of inter-regional system of magnetotelluric monitoring of seismoactive areas of Central Asia by means of Phoenix MTU-5D System 2000.

Stages of project implementation:

1. 1-st stage - delivery of two kits of measuring stations MTU-5D System 2000 to the Scientific Station OIVTRAN (Bishkek)
2. 2-nd stage –"pilot" geophysical investigations within the framework of the project on creating the inter-regional system of magnetotelluric monitoring of seismoactive areas of Central Asia by means of Phoenix MTU-5D System 2000 in 2003-2004.
3. 3-rd stage - long-term magnetotelluric (МТ) monitoring of seismoactive areas of Central Asia by means of Phoenix MTU-5D System 2000.

      The "pilot" stage of the project is primarily aimed at the effective performance of MT-monitoring system on the basis of the equipment Phoenix MTU-5D in seismogenerating zone of the northern Tien Shan in industrially advanced and densely populated area around the Bishkek geodynamic proving ground.

     By analyzing the received measurements, we can reveal certain areas and depths as the most sensitive elements of the non-equilibrium dynamic system, which is important for understanding of seismotectonic processes and their place in a stream of KNET data. Later on, we can consider such elements (zones) as potential indicators of changes in electromagnetic properties of the environment, which are related to the processes of the earthquake preparation, and they should receive a detailed study. Spatial localization of structural features, express-observations and the analysis of how these structures act in time will provide a good basis for the creation of the effective regional EM-monitoring network.

      Modern conceptions about seismotectonic processes strongly justify the use of observations of natural and artificial electromagnetic fields on the Earth’s surface for monitoring of geodynamic processes. Here are the two approaches: 1. Detection of electromagnetic signals generated by mechanoelectric and other transformations during the earthquake preparation; 2. Detection of changes in geoelectric cross-section forced by tectonic processes. The second approach can be brought into effect by the help of magnetotelluric methods. Till now, the given approach was basically restrained due to the low accuracy of obtaining the transfer functions of MT-field. The accuracy of parameter determination made 10-15 % at best. Such accuracy, for example, is commensurable with the value of maximal precursor anomalies of electrical resistivity detected by the help of controllable sources of electromagnetic field at Bishkek geodynamic proving ground [1].

      Since 1982, the Research Station RAS has been carrying out the operating electromagnetic observations based on the methods of vertical electrical sounding, frequency sounding and transient electromagnetic sounding (TEM) at a distant point at Bishkek geodynamic proving ground (150х100 km). The experiment has shown that, for all the period of observations, the maximal changes in electrical resistivity (from 5 up to 20 %) were usually observed before the earthquakes of class К>10, occurred in a controllable zone.

Fig. 1. The Bishkek geodynamic proving ground: allocation of stationary sites of electromagnetic monitoring TEM and experimental methodical magnetotelluric soundings by Phoenix MTU-5D. Relief is constructed on the basis of 1 km DEM.

The duration of anomalies varied at different sites and at different depths - from 20 days to 3 months, and the area made from 200 up to 400 км2. Thus, we have all the necessary preconditions for detecting the variations of electrical resistivity at different depths and for the estimation of their relation to seismotectonic process using modern technology of magnetotelluric sounding combined with TEM.
      Undoubtedly, the accuracy of measuring and determination of geophysical parameters is fundamentally important for the creation of a particular monitoring system, and it determines the effectiveness of magnetotelluric sounding for the study of seismotectonic processes in the earth's crust for the given research area. Additional optimism is given by the fact that "Phoenix Geophysics" company has developed a high-accuracy measuring system of new generation MTU-5D for continuous electromagnetic observations, allowing everyone to receive magnetotelluric transfer functions on the given site with the accuracy of 1-2 % [2].

The stability of the equipment response is extremely important for the stationary MT-system. Figure 2 shows the amplitude-phase frequency characteristics of magnetic channel MTU-5D with induction sensors MTS - 50. The three calibrations made at different time and various test checks of all measuring channels and sensors have shown the high stability of measuring system Phoenix MTU-5D. The changes in equipment characteristics have never exceeded 0.5 %.

      It is well known that the presence of noise in measurements can considerably bias the estimation of MT-parameters [3]. The method of reducing the mistakes of bias by using mutual correlation between two sites of synchronous recording is called the procedure of MT-observation with remote reference [4]. The procedure is based on synchronous measurements through two additional independent channels Rx and Ry. The basic channels are usually the measurements from the magnetic sensors located at a distance of several hundreds of meters up to 100-200 km from the sounding point, depending on the basic sources of noise and disturbance. To suppress the effect of instrumental noise, the distance of tens of meters is enough. For the territory of Bishkek geodynamic proving ground subject to a high level of industrial disturbances the basic point should be moved away not less than 50 km to exceed the radius of correlation of disturbing fields. If the noises are not correlated with useful signals, the estimations of MT-parameters received from the system of linear equations using basic channels, will be unbiased since the expressions include only mutual-spectral densities of capacity.

Figure 3 shows sounding curves for the chosen days at Tash-Bashat observation site. The top diagram shows the results of single processing, the bottom diagram shows the results of processing using the information from the remote reference Issyk-Ata. The figure clearly shows the obvious improvement in quality of curves as a result of remote reference processing, especially in a high-frequency band. The result shows the presence of strong local disturbances at Tash-Bashat site, whose influence on transfer functions was essentially lowered by the remote reference.

      What components of impedance tensor and what intervals of the periods should be used for monitoring the geoelectric cross-section? Let us consider the results of continuous three-day recording of MT-field at Kemin site (fig. 4). For the analysis of stability of determination of apparent resistance, we analyzed the values of apparent resistance received on the basis of two-hour sampling for the frequencies of 110; 11.25; 1.03125 Hz. After statistical processing, we found out that transfer functions are safely determined for the frequency of 11.25 Hz. On the given frequency, the expectation of apparent resistance rxy equals 161; root-mean-square deviation equals 4.96. For the phase Zxy, correspondingly, 37.06 and 0.969 degrees. A less stable determination of apparent resistance of 1.03125 Hz is probably explained by the fact that in a band of frequencies of 1-0.1 Hz the variations of MT-field are the least intensive.

Similar variability of apparent resistance estimations can be seen at Chon-Kurchak site (fig. 5). During 4 days we can see a clear period of about 24 hours that can reflect the modulation of electrical resistivity by tidal fluctuations (deformations) of lithosphere. The spectrum of lunisolar tide includes two (23.9 and 24.1 hours) and 12.0-hour components which define a daily course of tide [5].

Fig. 5. Time variability of apparent resistance module (. - rxy, + - ryx) and impedance phase (. - Zxy, + - Zyx) for the frequency of 11.25 Hz at Chon-Kurchak observation site. The beginning of recording – August 17, 2003, local time 7 o’clock p.m.

 It is important to study the influence of external factors (such as deformations of terrestrial (lithosphere) tides) on registered parameters, because the opportunity of detecting solid tides determines minimally allowable resolution threshold in the monitoring system from the viewpoint of estimating the influence of deformation processes. That is why we need an additional longer experiment and a quantitative estimation of relation of tidal deformations to the changes in electrical resistivity. It can finally confirm the importance of the given reason for the explanation of such short-period variations of resistance.

     In conclusion, it would be desirable to note that first practical results presented here give us optimism and confidence in high informative density of using electromagnetic fields of artificial and natural origin as a tool of monitoring seismotectonic processes at Bishkek geodynamic proving ground.

REFERENCES

1. Batalev V.Y., Bragin V.D., Zeigarnik V.A., Zubovich A.V., Matiks A.I., Orlenko N.N, Rybin A.K., Trapeznikov Y.A. i dr. – Monografia: Proyavleniye geodinamicheskih processov v geofizicheskih polyah – M., Nauka, 1993, 158 s (in Russian).
2. Remote reference research - automated MT station, Phoenix Geophysics Newsletters, Issue 20, p. 3. Website: www.phoenix-geophysics.com
3. Berdichevskiy M.N., Bezruk I.A., Safonov A.S. Magnetotelluricheskiye metody. Elektrorazvedka. Spravochnik geofizika. Moskva, Nedra, 1989, T.1.S. 261-310 (in Russian)
4. Goubau W.M.,Gamble T.D.,Clarke J. Error analysis for remote reference magnetotellurics // "Geophysics", 1979, v.44, N 5, 959-968.
5. Melhior P. Zemniye prilivy. M.: Mir, 1968 (in Russian)