Документ взят из кэша поисковой машины. Адрес оригинального документа : http://geophys.geol.msu.ru/shevnin/publ/128.pdf Дата изменения: Wed Oct 17 15:56:23 2012 Дата индексирования: Sat Feb 2 23:41:18 2013 Кодировка:

STREAM-FUNCTION USED FOR CURRENT-LINES' CONSTRUCTION IN 2-DIMENSIONAL DC MODELING
Bobachev A.A., Modin I.N., Pervago E.V., Shevnin V.A. Geological faculty, MSU, Moscow, 119899, Russia. Fax: (7-095)-9394963, E-mail: Boba@geophys.geol.msu.ru Internet site: www.geol.msu.ru/deps/geophys/rec_labe.htm Report, presented at the 5th Meeting EEGS-ES in Budapest, 5-9 September 1999. The stream function described is employed for the presentation of 2D DC modeling results. The 2D model is understood as a 2D medium with linear current electrodes, oriented along the inhomogeneities' strike direction. In this case both the medium and the electric field depend on two space coordinates only. Modeling becomes much easier than considering point current electrodes, where the electrical field always is three-dimensional. Meanwhile the actual results of such modeling are qualitatively equivalent to 3D modeling with point electrodes, as long as the measurements are conducted across the objects. The classical modeling presentation is in apparent resistivity which reflects an electric field distribution on the earth's surface. Quite often the connection of measured anomalies with a geoelectrical model is rather complex (fig. 1, A and C). The visualization of DC current lines simplifies understanding of the electric field's structure. Current lines are used in almost each textbook, but a practical technique for their construction is usually not included. The evident way for drawing current-lines is the step by step continuation of a line from some point along the electric field direction. The practical realization of such approach is not trivial. For a 2D field it is possible to use the stream-function. This function is often used in EM field modeling [flux function, Berdichevsky, 1984]. A contour map of the streamfunction corresponds to the stream-line distribution. Thus the problem of current streamlines' construction is reduced to the calculation of the streamfunction in the research area. This can be achieved by calculating secondary surface charges, which are determined at 2D modeling, using Fredholm's integral equation of the second type relatively of electric field [Escola, 1979]. The stream-function's () physical definition is the difference between stream-function's values in two points in space, which is equal to Fig.1. Apparent resistivity for a gradient array (A), DC current lines for central part of model (B) and the model the electric current intersecting a curve with the current electrodes' position (C). connecting them:

(B) - (A) = i(l ) n(l ) dl = (l ) E(l ) n(l ) dl ,
A A

B

B

(1)

where i is the vector of electric current density , n is the normal vector to the AB curve, E is the electric field intensity and (l) is the electric conductivity in a point l. The value of the integral (1) does not depend on the integration's path in regions without current sources. Therefore the stream-function can be calculated by integration from any point. Assuming that the stream-function is equal zero at point M, subsequent for any point K we can write an equation:

(K) = (l ) E(l ) n(l ) dl .
M

K

(2)

1


The electrical field is the sum of a primary field of current from the current electrodes and a secondary field of charges from the modeled body's surface (surface charges). If the density of surface charges is known from 2D modeling, the integral (2) can be calculated analytically. The connection of stream-line distribution with apparent resistivity ( a ) depends on the kind of array. For pole-dipole array with ideal MN dipole the following relation is valid and exact: j MN , (3) a = MN j0 MN and j MN are the resistivity and current density in MN center, j0 is the current density for a uniform half-space. Therefore, if the upper layer is homogenous ( MN = const ), then a is proportional to the current density. This approach clarifies the a physical explanation as shown in fig.1A. The current distribution around conductive object (fig.1B) results in changes of current density on the surface and this finds it's expression in a anomalies. a for pole-pole array is related to secondary charge distribution. An expression can be derived for the potential value in point M ( U M ): U M = U 0 + U S , (4) where U 0 is the potential of the primary electric field, U S is the potential of the secondary electric field, defined by the surface charge distribution. The surface charge density ( ) on the boundary of the inhomogeneity is determined from the boundary condition (Strattom 1941, p.163): 1 2 (5) = E n - E n = jn ( 1 - 2 ) . 0 Thus the surface charge density reveals a distribution of streamlines (fig.2). The upper left corner of the object is less expressed in the apparent resistivity than the upper right. Both corners will be expressed clearer in the case of a conductive object. The best way to visualize surface charge distributions, is the a cross-section for a polepole array (2B), corresponding to measurements with a buried potential electrode. Maxima and minima of such cross-section show a maximum density of positive and negative surface charges. In our practice stream-line function is used both for educational tasks and practical investigations. We can calculate on its base current lines, potential lines and apparent resistivity isolines in vertical cross-section. Like in formula 4 it is possible to separate potential, electric field and apparent resistivity into normal and anomalous parts Fig. 2. The model with highly resistive object and draw these lines distributions near anoma- ( =1000 m, body medium=100 m) for polelous object. Possibilities of this approach are pole array with buried current electrode. Aplisted in the table below and at some examples parent resistivity on the surface (A), in the shown at fig.3-10. cross-section (B) and DC stream-lines (C).

2


A 50 150 150

40

20 3. Example of a field's distortions, caused with high resistive object near measuring 4. Current lines' distribution near high resiselectrodes. (medium=30, body=100 Ohm.m, tive layer with conductive zone in three layXA=100). A - a -graph, B - current lines dis- ered model tribution in vertical cross-section.

5. A single current electrode near contact 6. A single current electrode near thin conwith conductive medium ductive layer

3


7. a graph as function r=AO (A), current lines and apparent resistivity distributions (B) for model with negative relief's form. Field technologies are shown on the right.

0

A

8. 2D modeling from current source near conductive object (Medium=100, body=15). A. Current lines and a vectors. B. Current lines of anomalous electric field outside of anomalous body and vectors of anomalous a.
100 1
40 30 20 40 30 20 10 0 -10 -20 -30 -40 -40

-10

10 0

-20

-10 -20

-30

-30 -40 -40

-30

-20

-10

0

10

20

30

-30

-20

-10

0

10

20

30

10. DC current and iso-potential lines from -30 -20 -10 0 10 20 30 40 9. DC current and iso-potential lines near in- point current source near high and low resistive cylinders clined high resistive body
-40 -40

For streamlines calculation DC_flow software has been developed. Working version of DC_flow software you can find in Internet at our web-site. References
Berdichevsky M.N. and Zhdanov M.S. 1984, Advanced Theory Of Deep Geomagnetic Sounding, ELSEVIER, Amsterdam. Escola, L. 1979, Calculation of galvanic effects by means of the method of sub-areas, Geophysical Prospecting 27, 616-627. Stratton J.A. 1941, Electromagnetic Theory, McGraw-Hill, New York.

4