CA1246673A - Electromagnetic array profiling survey method - Google Patents
Electromagnetic array profiling survey methodInfo
- Publication number
- CA1246673A CA1246673A CA000508778A CA508778A CA1246673A CA 1246673 A CA1246673 A CA 1246673A CA 000508778 A CA000508778 A CA 000508778A CA 508778 A CA508778 A CA 508778A CA 1246673 A CA1246673 A CA 1246673A
- Authority
- CA
- Canada
- Prior art keywords
- earth
- survey
- variations
- measured
- survey line
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired
Links
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
ELECTROMAGNETIC ARRAY PROFILING SURVEY METHOD
ABSTRACT
An electromagnetic survey method for geophysical exploration, in which the variations in the earth's magnetic field are measured in two, non-parallel directions at one point in the survey area.
Simultaneously, the variations in the earth's electrical field parallel to the survey line are measured at a number of points along the survey line. These measured variations are transformed to the frequency domain, and then the horizontal component of the magnetic field orthogonal to the direction of the measured electrical field is calculated. The impedance at each measurement point on the survey line is calculated as a function of frequency, and weighted averages of the impedances for predetermined frequencies using a zero phase length weight function corresponding to a low pass filter applied to the electric field are used to calculate the subsurface conductivity distribution.
ABSTRACT
An electromagnetic survey method for geophysical exploration, in which the variations in the earth's magnetic field are measured in two, non-parallel directions at one point in the survey area.
Simultaneously, the variations in the earth's electrical field parallel to the survey line are measured at a number of points along the survey line. These measured variations are transformed to the frequency domain, and then the horizontal component of the magnetic field orthogonal to the direction of the measured electrical field is calculated. The impedance at each measurement point on the survey line is calculated as a function of frequency, and weighted averages of the impedances for predetermined frequencies using a zero phase length weight function corresponding to a low pass filter applied to the electric field are used to calculate the subsurface conductivity distribution.
Description
UTSB:210 ~2~Çi,673 ., ELECTROMAGNETIC ARRAY PROFILING SURVEY METHOD
This invention relates to the field of geophysical exploration using measurements of naturally occurring or artificially induced geomagnetic fields and induced geoelectric (telluric) fields, collectively called magnetotelluric fields.
The flow of the telluric currents through the earth's crust depends upon the conductivity or resistivity o the structure of the crust at any particular point. If this conductivity or resistivity can be measured and mapped, information about the structur~, and in particular as it may relate to the presence of hydrocarbon, mineral or geothermal resources, can ~e gained. This is particularly useful in areas where seismic methods of geophysical surveying cannot be used, such as where the sedimentary rGcks are overlaid with a thick volcanic layer.
The telluric currents constantly vary in maqnitude, - direction and polarity, and constitute a complex spectrum of components with various frequencies.
::
:
This invention relates to the field of geophysical exploration using measurements of naturally occurring or artificially induced geomagnetic fields and induced geoelectric (telluric) fields, collectively called magnetotelluric fields.
The flow of the telluric currents through the earth's crust depends upon the conductivity or resistivity o the structure of the crust at any particular point. If this conductivity or resistivity can be measured and mapped, information about the structur~, and in particular as it may relate to the presence of hydrocarbon, mineral or geothermal resources, can ~e gained. This is particularly useful in areas where seismic methods of geophysical surveying cannot be used, such as where the sedimentary rGcks are overlaid with a thick volcanic layer.
The telluric currents constantly vary in maqnitude, - direction and polarity, and constitute a complex spectrum of components with various frequencies.
::
:
-2~ 6673 The first practicable electromagnetic survey method was described by Cagniard in U. S. No. 2,677,801. This consists of measuring and recording over a period of time the variations in one horizontal componant of the telluric field, and simultaneously measuring the recording the orthogonal component of the yeomagnetic field. These measurements are then converted into frequency components by means of Fourier analysis. The ratio of the frequency component of the electrical field to that of the magnetic field is a wave impedance that is a function of frequency.
Since the depth of penetration of an electromagnetic wave into the earth is inversely related to the frequency of the wave and the conductivity of the earth, the wave impedance can be used to estimate the conductivity distribution in a vertical direction through the earth's surface. Cagniard made this estimate using a mathematical model in which conductivity varied only with depth, the so-called l-D model.
This method was subsequently developed by other workers to be used with models in which the conductivity varies with one horizontal coordinate as well as with depth, the 2-D model. The coordinate along which conduc-tivity is constant in this model is called "strike". In this model there are two cases to be considered, based on the polarization of the magnetotelluric fields. In these two cases the electric fields are polarized respectively parallel and perpendicular to strike. Theoretical studies show that the E parallel impedance function can be used with the techniques developed for l~D to give reasonably accurate estimates of the conductivity directly beneath the site at which khe function is computed. However, the E perpendicular impedance functions have very different properties, and cannot be directly inverted to produce an estimate of conductivity which is o~f reason~ble accuracy.
The most common solution of this problem is to use the , .~
, _3_ ~ 73 curves obtained at more than one site, and to adjust the cond~ctivity distribution in the model by an iterative process until theoretical E perpendicular curves which best fit the observed results are obtained. This method has many drawbacks, in particular it requires a consider-able amount of computational effort, and there is no assurance that the solution obtained is in fact the correct one.
These methods work well when the area being sur~eyed does have conductivity which varies only in one or two directions. Unfortunately, however, such two-dimensional variation is rare. When magnetotelluric measurements are made over a structure having a three dimensional conduc-tivity distribution, the following problems are encoun-tered. First, it is generally not possible to identiy a set of principal axes as it is in the 2-D case, although a number of ad hoc procedures exist to generate a set of axes. Second, whatever coordinate system is used, it is not possible to separate the electric field into the two distinct cases with the distinctive properties of E
parallel and E perpendicular found in the 2-D case. The desirable properties of the E parallel case disappear first, and both cases take on properties similar to those of E perpendicular. Various ad hoc procedures were developed to try and deal with this problem, but they have had limited success. A third problem is that the considerable additional complexity significantly incraases the difficulty of ensuring that the interpretation of the
Since the depth of penetration of an electromagnetic wave into the earth is inversely related to the frequency of the wave and the conductivity of the earth, the wave impedance can be used to estimate the conductivity distribution in a vertical direction through the earth's surface. Cagniard made this estimate using a mathematical model in which conductivity varied only with depth, the so-called l-D model.
This method was subsequently developed by other workers to be used with models in which the conductivity varies with one horizontal coordinate as well as with depth, the 2-D model. The coordinate along which conduc-tivity is constant in this model is called "strike". In this model there are two cases to be considered, based on the polarization of the magnetotelluric fields. In these two cases the electric fields are polarized respectively parallel and perpendicular to strike. Theoretical studies show that the E parallel impedance function can be used with the techniques developed for l~D to give reasonably accurate estimates of the conductivity directly beneath the site at which khe function is computed. However, the E perpendicular impedance functions have very different properties, and cannot be directly inverted to produce an estimate of conductivity which is o~f reason~ble accuracy.
The most common solution of this problem is to use the , .~
, _3_ ~ 73 curves obtained at more than one site, and to adjust the cond~ctivity distribution in the model by an iterative process until theoretical E perpendicular curves which best fit the observed results are obtained. This method has many drawbacks, in particular it requires a consider-able amount of computational effort, and there is no assurance that the solution obtained is in fact the correct one.
These methods work well when the area being sur~eyed does have conductivity which varies only in one or two directions. Unfortunately, however, such two-dimensional variation is rare. When magnetotelluric measurements are made over a structure having a three dimensional conduc-tivity distribution, the following problems are encoun-tered. First, it is generally not possible to identiy a set of principal axes as it is in the 2-D case, although a number of ad hoc procedures exist to generate a set of axes. Second, whatever coordinate system is used, it is not possible to separate the electric field into the two distinct cases with the distinctive properties of E
parallel and E perpendicular found in the 2-D case. The desirable properties of the E parallel case disappear first, and both cases take on properties similar to those of E perpendicular. Various ad hoc procedures were developed to try and deal with this problem, but they have had limited success. A third problem is that the considerable additional complexity significantly incraases the difficulty of ensuring that the interpretation of the
3~ data obtained is in fact the correct one.
One way of reducing the third problem is to take many more measurements. In the classical magnetotelluric method, it is usually more difficult to measure the magnetic field than it is to measure the electrical~field, making a large number of measurements time-consuming and .~ .
~4~ 7~
expensive. However, it is an observed fact that the horizontal components of the magnetic field typically vary much more slowly over distance than do the horizontal components o the electric field. This gave rise to the magnetotelluric-telluric method, which involves the measurement of two orthogonal components of the magnetic field at a limited number of sites distributed over the survey area, and measurement of two orthogonal components of the electrical field at many more sites. The data are then processed using techniques similar to those described for the magnetotelluric model. ~owever, when the conductivity distribution is three dimensional, the results are still found to be fairly unreliable. One variant of this method, which uses a method equivalent to the "roll-along" method of seismic geophysical exploration, is described in U. S. No. 4,286,218.
Overlapping measurements of the electrical fields along the survey line improve the signal to noise ratio, an~ the magnetic field is measured at either end of the line, the intermediate magnetic field at the points at which the electrical field is measured being derived by interpolation.
The main disadvantage of all these conventional methods is that they give very unreliable results when the electrical and magnetic characteristics of the subsurface structure vary in all three dimensions. Prior attempts to overcome this unreliability have involved huge computa-tional efforts, which are rarely justified in terms of the 3~ results produced.
Another disadvantage of conventional magnetotelluric survey methods is that two measurements of the electrical field, preferably in orthogonal dire~tions, are required at each survey point. This makes the use of this method particularly difficult in offshore surveying. An offshore _5~ 73 method for a magnetotelluric survey is describecl in U. S.
No. 4,210,869, but this method relies upon acoustic methods of measuring the position and orientation of the electrodes on the sea floor. The accuracy of these measurements is limited, making the interpretation of the survey results unreliablP.
A further disadvantage of conventional magnetotel-luric methods is that measurements of the magnetic field at several points in the survey area is re~ired, unless the area to be surveyed is small. In addition, the verti-cal component must be ascertained at each E measurement point in order to distinguish between the apparent E
parallel and E perpendicular components. As explained above, if this distinction is not made, there is a serious likelihood that errors will be made in interpreting the data obtained.
The above noted and other disadvantages of the prior art are overcome by providing an electromagnetic geophysical survey method which gives reliable results in the presence of three dimensional variation of the conductivity of the earth's structure at the survey points by means of relatively simple computational methods.
According to one aspect of the invention, the elec-tromagnetic geophysical survey method comprises measuring the variations in the earth's magnetic field in at least ~ two horizontal non-parallel directions at at least one point in the survey area and simultaneously with the meas-urement of the variations of the magnetic field, measuring the variations in the earth's electrical field parallel to a survey line at a plurality o points along the survey line. The measured data are then transformed into frequen2y components. The next steps are calculating as a function of frequency the horizontal component of the . .
earth's magnetic field orthogonal to the direction of the measured electrical field at each of the points from the measurements of the magnetic field in the two non-parallel directions, calculating as a unction of frequency the S impedance at each point, the impedance being the ratio between the measured electrical field at that point and the horizontal component of the earth's magnetic field orthogonal to the direction of the measured electrical field. Then, for predetermined frequencies, the weighted averages of the impedances are calculated such that the number of impedances entered into each weighted average increases with decreasing freguency in such a way that the number is substantially proportional to the effective depth of penetration into the earth of an electromagnetic wave of that frequency. Einally, the distribution of conductivity in the earth below the survey line is calculated as a function of depth from the wei~hted averages of the impedances.
In one embodiment, the two directions in which the earth's magnetic field is measured are orthogonal to one another.
In another embodiment, the step of measuring the variations in the earth's electrical field parallel to the survey line consists of measuring the variations in the potential differences between one or more pairs of elec-trodes in electrical contact with the earth and spaced at substantially equal distances along the survey line and dividing the measured potential differences by the distance between the electrodes.
.
In a further embodiment, the variations in the poten-tial differences between all the electrodes are measured simultaneously.
.
In an alternative embodiment, the variations in the potential differences between the members of a group of adjacent electrodes consisting of less than all the elec-trodes are measured simultaneously and the measurements for each of the groups are made sequentially.
In another alternative embodiment, at least one of the electrodes in one group is also a member of the adja-cent group.
In the preferred embodiment, the weighted averages of the impedance~ are calculated using a zero phase finite length weight function, the width of the weight function for each frequency being determined by selecting an appropriate width, obtaining a weighted average impedance using the selected function, using the weighted average so obtained to calculate an apparent depth of penetration, comparing the calculated depth of penetration to the expected depth of penetration, using the difference between the calculated and expected depths to assist in the selection of a more appropr-ate width, and repeating the process iteratively until a predetermined accuracy of calculation of the depth of penetration is obtained.
According to another aspect of the invention, the electromagnetic geophysical survey method consists of measuring the variations in the earth's magnetic fiel~ in at least two horizontal non-parallel directions at at least one point in the survey area, simultaneously with the measurement of the variation of the magnetic ~ield, `:
measuring the variations in the potential differences between ad;acent electrodes in electrical contact with the earth, the electrodes being spaced at substantially equal intervals along a survey line and the measurements of the potential differences being made simultaneously and trans-forming the measured~variations in the potential differ--8~ 73 ences to a function of frequency. There is then calcula-ted for predetermined frequencies the weighted averages of the measured potential differences such that the number of potential differences entered into each weighted average increases with decreasing frequency in such a way that the number is substant.ially proportional to the effective depth of penetration into the earth of an electromag~etic wave of that frequency. The horizontal component of the earth's magnetic field orthogonal to the direction of the survey line is calculated from the measurements of the magnetic field in the two non-parallel directions, the weighted impedances along the survey line are calculated from the ratio of the weighted averages of the potential differences and the horizontal orthogonal component of the magnetic field and the distribution of conductivity of the earth below the survey line is calculated as a function of depth from the weighted impedances.
According to yet another aspect of the invention, the electromagnetic geophysical survey method comprises addi-tionally the steps of measuring the vertical component of the earth's magnetic field at a plurality of points along the survey line and using these measured vertical magnetic components together with the calculated distribution of conductivity of the earth below the survey line to deter-mine the geophysical structure of the earth below the survey line.
It is an advantage of the invention that it provides a new and improved method of estilnating the conductivity distribution of the earth's substructure below a survey line when that conductivity distribution varies in all three dimensions. Furthsr advantages of the invention are ~to bs found in the simplicity of the reguired placing of the electrical and magnetic field measuring devicss in tha ~ survey method. A ~urther advantage is that this invention ,~
.~, 7~
g requires simpler computational efforts to produce esti-mates of complex conductivity distributions than were required by prior methods.
The above, and other objects and advantages of the present invention will bPcome more apparent from a detailed description of the preferred embodiments when read in conjunction with the drawings.
lQ FIGURE 1 is a diagrammatic representation of the survey field.
The field layout for a survey in accordance with the present invention is shown in Figure 1. A suitable site in the survey area for the magnetic field measurements is selected, preferably rernote from localized non-natural sources of electromagnetic disturbance such as generators, high tension power lines, radio transmitters and busy roads. Magnetic field sensors 10 and 12, of a convention-al type, are installed so as to measure the fluctuationsof the horizontal component of the earth's magnetic field in two non-parallel directions. In the preferred embodiment, these sensors are at right angles to each other, ~ith one sensor parallel to the proposed survey line. Readings from the magnetic field sensors 10 and 12 may be transmitted to a processor station 14 in real time by radio telemetry, as indicated in FIGURE 1, or by direct cable transmission or over telephone lines.
In an aIternative embodiment, the data is recorded in a storage medium such as an electronic computer memory~
magnetic tape or disc resident at the recording site. The stored data could either be transmitted to the processor stations in delayed time or the storage medium containing ~; 35 the data could be phys~ically transported to the processor ~,.
3L2~6~7~
station. This allows independent, unattended operation of the recording stations.
While a single magnetic measurement station is pre-ferred as giving coherent reference signals, if circum-stances demand more than one magnetic station location may be used. If it is necessary as the survey progresses to move the magnetic station, a second station with two sen-sors similarly aligned to the first is set up at the new position, and simultaneous measurements are made for a suitable period in order to translate data obtained using the second station to the same frame of reference as data obtained using the first station, a procedure familiar to those skilled in the art.
A survey line 16 is then determined. This line is preferably straight, but the method may also be used with curved survey lines. Electrodes 18 are installed in elec-trical contact with the earth along the survey line, pref-erably spaced e~uidistant from each other. Adjacent elec-trodes are connected by wires to a potential difference sensor 20. Readings from the potential difference sensors 20 are transmitted to the processor station 14 by radio, cable or telephone or they may be recorded on a mass storage medium at the recording site for subsequent transfer to the processor station, as in the case of the magnetic readings.
The E (electrical) field component is o~tained by dividing the measured potential difference between two electrodes by the electrode spacing L. This type of E
~ield measurement is in ~act a weighted average of the E
~ield component along the wire connecting the two elec-trodes. The arrangement therefore acts as a spatial low pass~filter in wave number space, with an inherent weight-ng factor of a rectangle or boxcar function. The ...
response of the sensor in wave number space is proportion-al to a sine (sinx/x) function with a cutoff wave number somewhat less than the reciprocal of 1,. Those skilled in the art will therefore be able to select a suitable value of L so as to provide the desired survey rasolution.
A se~uence of component E field measurements is made along the survey line 16. As shown in Figure 1, the sen-sors are installed end to end. This arrangement spatially low pass filters the continuous E field component along the path of the survey, then samples this filtered compon-ent at the regular spatial interval L. This type of -sampling is consistent with the requirements of the well-known Sampling Theorem necessary to prevent aliasing of the sampled data. The length of the spread of measure-ments on either side of the points for which it is desired to obtain conductivity data should be approximately equal to the maximum depth to be investigated. The E field measurements may be made simultaneously over the whole survey line or they may be made at one time over one group of adjacent electrodes and then at a later time over another group until the whole survey line has been covPred in se~uence. This latter method is advantageous when surveying to great depth so that the required spread of measurements is very lon~, and simultaneous measurements over the whole line would therefore be impracticable.
When sequential measurements are made, a small overlap in the membership of adjacent groups may be used to provide redundancy to assist in the reduction of noise, but this slows down the survey so preferably it is only used when noise is a serious problem.
When the survey line 16 is curved the process must be modified. First, a low pass filter in the form of a 35 ~weighted sum is~applied to the impedances describing the relationships between the components of the electric field .
i;73 - ~2 -tangential to each of a group of adjacent path se~ments and the component of the magnetic field perpendicular to an effective direction defined for th~ group of segments.
The simple cosine bell weighting function is also used for a curved survey line.
The effective direction of the line segments at a point ~ locating the center of the seg~lent group is deter-mined by defining the angle eo(~) between the effective direction of the group and an arbitrary reference direction, say north. The defining relationship for the angle is N2 (CY) ~ Wi(~)SIN~i i=Nl (CY) eot~) = tan~~ _ N2 (~) ~ Wi (~) COS~i i=N1(~) where the Wi(~) are the cosine bell weights, N1(~) and N2(~) are the lower and upper limits for the sum required to include all of khe non-zero weights over the length of the cosine bell and the ~i are the angles between the survey line se~ments and the reference direction.
The sum of the segment impedances weighted by the cosine bell is then used to compute an apparent depth of penetration. Also, an effective length~ LE(~)l for the survPy line segments in the vicinity of the point a is computed from the defining relation N2 (~) I, ~ Wi(~)COS~o(~
LE(~)= W~ax i=N
`~' -~LZ~ 7~
where L is the length of the individual path segments, all of which are assumed to be equal in length, and WmaX is the largest one of the Wi weights included in the sum.
For the case of thP curved survey path the effective length, LE(~), rather than the length of the cosine bell fun~tion, is required to be a specified multiple of the apparent depth of penetration.
The required value o~ LE(~) is obtained by a process o~ iteration. The process begins by making an initial guess of a suitable length for the cosine bell function.
The weights, Wi(~) are then computed and used to obtain the angle ~O(~) and the effective length LE(~. The apparent depth of penetration is computed from the weighted sum o~ the segment impedances. The effective length and the depth of penetration ara comparecl and the dif~erence between the computed and specified multiple used to determine a new length for the cosine bell. ~he process is repeated in iterative fashion until a suitable measure of convergence is achieved. The procedure is used at each frequency and position on the survey line 16, to be examined.
In the preferred embodiment impedances which are the ratio o~ the fre~uency components of the E field measurement to those o~ the magnetic field mPasured simultaneously in the orthogonal direction to the E ~ield ara then calculated. This impedance designated Zxy is an element of a rank two impedance tensor relationship between the spatial components of the horizontal el~ctric and magnetiç fields. Zxy is defined by the equation Ex = ZxxHx ~ Zx~H~
where Hx is the transform of the magnetic field measure~
ment parallel to the E field measurement, an~ Hy is the transform of the magnetic field mea~uremen~ orthogQnal to the ~ ~ield. The impedance Zxy may be estimated from -14~ 73 the recorded data and the defining equation by any one of a variety of techniques familiar to those skilled i.n the prior art of magnetotelluriçs.
In an alternative embodiment of this invention, the electrical field may also be measured in the direction at right angles to the survey line. In such a case another element of the impedance tensor designated Zxy and defined by the equation Ey = ZyxHx + ZyyHy may also be estimated from the frequency components of the measured data.
It has been discovered as part of this invention that the function Zxy in three dimensional analysis behaves similarly to the E parallel impedance function in conven-tional two dimensional magnetotelluric analysis, except at large wave number values. In this invention the desirable qualities of the E parallel function are obtained in the case of three dimensional conductivlty variation in the earth by attenuating the dependence of the electric field as reflected in Zxy at large wave number values by low pass filtering with respect to wave number space.
It has further bee.n discovered that a critical factor of the low pass filtering operation is the value of the cutoff wave number. It has bee.n determined that, with respect to increasing values of wave number, Zxy and the 2-D E parallel function begin to differ appreciably at the value of the wave length equal to the depth of penetration ~; ~ of an electromagnetic wave into the earth, therefore the reciprocal of this value is an appropriate cutoff, although other values may be used within the scope of the inventi OIl .
, :
, It is well known in the art that, in the wave number and space domains, the linear operation that corresponds to low pass Eiltering is convolution with a kernel weight-ing function. The spatial length of the kernel function is reciprocal with the cutoff wave number so that the length of the kernel increases with decreasing wave number. In order to perform the weighting operation adequately at all wave numbers, it is therefore necessary to have enough spread of E field measuremsnts at each point at which a conductivity sounding is to be obtained so that the longest kernel function required to spatially low pass filter the Zxy impedances can be accommodated.
In the preferred embodiment, the simple cosine bell function may be used as the weighting kernel function.
The exact length of the cosine bell function to be used at any one frequency at any position along the survey line 16 is determined in the preferred embodiment by an iterative method. At each frequency and position to be examined an initial guess of a suitable length of the cosine bell function is made. The impedances are then weighted by that function and the sum of the impedances used to compute an apparent depth of penetration. The length of the cosine function is required to be a specified multiple of this depth of penetration. The preferred embodiment uses a multiple equal to one. The length of the cosine bell is compared to the depth o:E
penetration and the diEference between the computed and specified multiple used to determine a new length for the cosine bell. This process is repeated until a suitable measure of convergence of the computed to the speciied multiple is obtained. In practice this iterative procedure has been found to work well and converge rapidly.
, .
:' -16~ 73 When all the E field measurements over the entire survey path are made simultaneously, an alternative method of obtaining weighted impedances may be used. The weight~
ing procedure, ayain using a cosine bell function of suit-able lenyth, is performed on the E field measurementstransformed to the fre~uenc y domain, and those weighted measurements are then used in the impedance tensor to obtain weighted Zx~ values.
Once the impedances have been spatially weighted at all frequencies of interest the functions are inverted to obtain estimates of conductivity versus depth along the survey path. The inversion procedures used in the preferred embodiment are those used in the prior art one dimension analysis. It would be possible to use more complicated procedures in cases where the increase in accuracy obtained by the use of such procedures justifies the considerable increase in computational effort required by their use.
The method by which the various computations are made is not part of the present invention, but it will be obvious to those skilled in the art that such computations may be conveniently made ~y use of a programmed digital electronic computer. Programs for performing the Fourier transforms, the cosine bell weighting function and the one d1mension inversion procedure are all readily available.
The estimated conductivity at any point in a survey made according to the method of the present invention is a weighted a~erage of the subsurface conductivities both directly underneath, and lateral of, the survey line.
When the bulk of a conductive feature lies lateral of the survey line but is still evident in the cross saction the section is said to be "side-swiped". According to pub-lished theoretical studies, the vertical component of the -17~ 67~
magnetic field at the point of the survey may be used to diagnose a side swipe. Therefore, in an alternative embodiment of this invention, the vertical component of the ma~netic field is measured at the points along the survey path, and these measurements used according to the published methods of analysis to determine whether the cross sections determined by the method described above contain side swipes.
In situations where unusual inhomogeneity of the magnetic field is encountered or anticipated, or where the environment is unusually noisy, or where a higher degree of accuracy is required, additional measurements of the magnetic field in two non-parallel horizontal directions can be made. These measurements may be made along the survey line, but more preferably at points througnout the survey area. The two directions are preferably orthogonal, with one direction preferably being parallel to the survey line.
If the x and y components of the magnetic field are measured at N separate locations, either simultaneously or at different times, the impedances Zax and Zay to be estimated for each span of the electric field array, a span being the distance between two adjacent electrodes, are assumed to satisfy the expression Ea - Zax ( ~ Ui Hxi) + Zay ~ Vi Hyl) where Ui and Vi are appropria~e weighting factors which may be determined by Yarious means familiar to those skilled in the art.
Designating the x and y field components at the reference site Rx and Ry, these field components being electric, magnetic or a combination of the two, the . !.11 -18~ 3 relationship to the x and y magnetic components at the ith site, Hxi and Hyi, is given by Hxi = Txxi Rx + Txyi Ry Hyi = Tyxi Rx ~ Tyyi Ry The transfer ratios, Txii, Txyi, Tyxi and Tyyi may be determined by any of the known referencing techniques conventionally used in magnetotelluric surveying.
Substituting these expression into the equation for Ea gives Ea = Zax ~J (Vi Txwi ~ Vi Tyxi) Rx + Zay ~Vi Txyi + Vi Tyyi) Ry From this Zax and 2ay may be obtained from the measured values of Ea, Rx and Ry by conventional means. The impedances thus determined for each span of the electric field array are then filtered by a process equivalent to convolution with a kernel weighting function as described above. The earth conductivity distributions calculated in this way should be more accurate than those involving single site magnetic measurements, although the additional magnetic measurements would make the survey more expensive to perform.
The embodiments of the invention described ahove relate to survey methods using alternating currents, whether naturally or artificially induced. The invention may also be applied to electrical surveying using a direct current, controlled source of excitation. In this type of surveying, the apparent depth of penetration into the earth of the signals is related to the distance be~ween the source and the point at which the electric field is measured. An electric field is excited in the earth by a :
:
:, -19~ 3 direct current controlled source, techniques for which are well known to those skilled in the art. The excited electric field is measured along a survey line at a distance from the source by the technique described above, namely measuring the potential diference between each pair of adjacent electrodes in the array, and dividing the measured potential difference by the spacing between the electrodes. The measured electric field data is then converted to apparent resistances, computed as the ratio of the measured electric field to the current put out by the controlled source, using conventional processing techniques. Weighted averages of the apparent r~sistances for predetermined distances from the source are obtained in such a way that the process is e~uivalent to using the low pass filtering technique utilizing a kernel weighting function on the electric field along the survey path, as described above. In this case the spatial length o~ the kernel function is reciprocal with a cutoff wave number which is inversely related to the distance between the source and the electric field measurement array.
It is an advantage of the present invention that it permits use of a much simpler configuration o electric field sensors than was possible with the prior magnetotel-luric survey methods. This will be of particular benefitin offshore surveying, and in surveying in very difficult terrain. Another advantage, which will be of benefit in the same situations, is that the magnetic field may be measured in a much simpler manner. Only one magnetic sensor site is required, which in offshore surveying could be on shore, for surveys near the coastline, or positioned adjacent to a fixed offshore platform. Further, unlike conventional magnetotelluric methods, measurement of the ; vertical component of the magnetic field is not required ~to make the suFvey although it may optionally be made to .~ .
, ' ' . .
~ 73 add information relating to structures lateral of the survey line.
It is a further advantage of $his invention that it can be readily adapted to a wide range of survey applications. Without in any way limiting the scope of the invention herein claimed and disclosed, it may be used in prospecting for minerals and hydrocarbons, in locating underground pipes and utility conduits, in detecting seepage of radioactive or toxic materials from storage sites, in detecting underground water supplies, in archeological research and in probing the conductivity of tissue, bones and organs within the body.
Additional advantages and modifications will be readily apparent to those skilled in the art. The inven-tion in its broader aspects is therafore not limited to the specific details, representative apparatus or the illustrative example shown and described. Accordingly, departures may be made from the detail without departing from the spirit or scope of the disclosed general inven-; tive concept.
:: `
:
1~
.
One way of reducing the third problem is to take many more measurements. In the classical magnetotelluric method, it is usually more difficult to measure the magnetic field than it is to measure the electrical~field, making a large number of measurements time-consuming and .~ .
~4~ 7~
expensive. However, it is an observed fact that the horizontal components of the magnetic field typically vary much more slowly over distance than do the horizontal components o the electric field. This gave rise to the magnetotelluric-telluric method, which involves the measurement of two orthogonal components of the magnetic field at a limited number of sites distributed over the survey area, and measurement of two orthogonal components of the electrical field at many more sites. The data are then processed using techniques similar to those described for the magnetotelluric model. ~owever, when the conductivity distribution is three dimensional, the results are still found to be fairly unreliable. One variant of this method, which uses a method equivalent to the "roll-along" method of seismic geophysical exploration, is described in U. S. No. 4,286,218.
Overlapping measurements of the electrical fields along the survey line improve the signal to noise ratio, an~ the magnetic field is measured at either end of the line, the intermediate magnetic field at the points at which the electrical field is measured being derived by interpolation.
The main disadvantage of all these conventional methods is that they give very unreliable results when the electrical and magnetic characteristics of the subsurface structure vary in all three dimensions. Prior attempts to overcome this unreliability have involved huge computa-tional efforts, which are rarely justified in terms of the 3~ results produced.
Another disadvantage of conventional magnetotelluric survey methods is that two measurements of the electrical field, preferably in orthogonal dire~tions, are required at each survey point. This makes the use of this method particularly difficult in offshore surveying. An offshore _5~ 73 method for a magnetotelluric survey is describecl in U. S.
No. 4,210,869, but this method relies upon acoustic methods of measuring the position and orientation of the electrodes on the sea floor. The accuracy of these measurements is limited, making the interpretation of the survey results unreliablP.
A further disadvantage of conventional magnetotel-luric methods is that measurements of the magnetic field at several points in the survey area is re~ired, unless the area to be surveyed is small. In addition, the verti-cal component must be ascertained at each E measurement point in order to distinguish between the apparent E
parallel and E perpendicular components. As explained above, if this distinction is not made, there is a serious likelihood that errors will be made in interpreting the data obtained.
The above noted and other disadvantages of the prior art are overcome by providing an electromagnetic geophysical survey method which gives reliable results in the presence of three dimensional variation of the conductivity of the earth's structure at the survey points by means of relatively simple computational methods.
According to one aspect of the invention, the elec-tromagnetic geophysical survey method comprises measuring the variations in the earth's magnetic field in at least ~ two horizontal non-parallel directions at at least one point in the survey area and simultaneously with the meas-urement of the variations of the magnetic field, measuring the variations in the earth's electrical field parallel to a survey line at a plurality o points along the survey line. The measured data are then transformed into frequen2y components. The next steps are calculating as a function of frequency the horizontal component of the . .
earth's magnetic field orthogonal to the direction of the measured electrical field at each of the points from the measurements of the magnetic field in the two non-parallel directions, calculating as a unction of frequency the S impedance at each point, the impedance being the ratio between the measured electrical field at that point and the horizontal component of the earth's magnetic field orthogonal to the direction of the measured electrical field. Then, for predetermined frequencies, the weighted averages of the impedances are calculated such that the number of impedances entered into each weighted average increases with decreasing freguency in such a way that the number is substantially proportional to the effective depth of penetration into the earth of an electromagnetic wave of that frequency. Einally, the distribution of conductivity in the earth below the survey line is calculated as a function of depth from the wei~hted averages of the impedances.
In one embodiment, the two directions in which the earth's magnetic field is measured are orthogonal to one another.
In another embodiment, the step of measuring the variations in the earth's electrical field parallel to the survey line consists of measuring the variations in the potential differences between one or more pairs of elec-trodes in electrical contact with the earth and spaced at substantially equal distances along the survey line and dividing the measured potential differences by the distance between the electrodes.
.
In a further embodiment, the variations in the poten-tial differences between all the electrodes are measured simultaneously.
.
In an alternative embodiment, the variations in the potential differences between the members of a group of adjacent electrodes consisting of less than all the elec-trodes are measured simultaneously and the measurements for each of the groups are made sequentially.
In another alternative embodiment, at least one of the electrodes in one group is also a member of the adja-cent group.
In the preferred embodiment, the weighted averages of the impedance~ are calculated using a zero phase finite length weight function, the width of the weight function for each frequency being determined by selecting an appropriate width, obtaining a weighted average impedance using the selected function, using the weighted average so obtained to calculate an apparent depth of penetration, comparing the calculated depth of penetration to the expected depth of penetration, using the difference between the calculated and expected depths to assist in the selection of a more appropr-ate width, and repeating the process iteratively until a predetermined accuracy of calculation of the depth of penetration is obtained.
According to another aspect of the invention, the electromagnetic geophysical survey method consists of measuring the variations in the earth's magnetic fiel~ in at least two horizontal non-parallel directions at at least one point in the survey area, simultaneously with the measurement of the variation of the magnetic ~ield, `:
measuring the variations in the potential differences between ad;acent electrodes in electrical contact with the earth, the electrodes being spaced at substantially equal intervals along a survey line and the measurements of the potential differences being made simultaneously and trans-forming the measured~variations in the potential differ--8~ 73 ences to a function of frequency. There is then calcula-ted for predetermined frequencies the weighted averages of the measured potential differences such that the number of potential differences entered into each weighted average increases with decreasing frequency in such a way that the number is substant.ially proportional to the effective depth of penetration into the earth of an electromag~etic wave of that frequency. The horizontal component of the earth's magnetic field orthogonal to the direction of the survey line is calculated from the measurements of the magnetic field in the two non-parallel directions, the weighted impedances along the survey line are calculated from the ratio of the weighted averages of the potential differences and the horizontal orthogonal component of the magnetic field and the distribution of conductivity of the earth below the survey line is calculated as a function of depth from the weighted impedances.
According to yet another aspect of the invention, the electromagnetic geophysical survey method comprises addi-tionally the steps of measuring the vertical component of the earth's magnetic field at a plurality of points along the survey line and using these measured vertical magnetic components together with the calculated distribution of conductivity of the earth below the survey line to deter-mine the geophysical structure of the earth below the survey line.
It is an advantage of the invention that it provides a new and improved method of estilnating the conductivity distribution of the earth's substructure below a survey line when that conductivity distribution varies in all three dimensions. Furthsr advantages of the invention are ~to bs found in the simplicity of the reguired placing of the electrical and magnetic field measuring devicss in tha ~ survey method. A ~urther advantage is that this invention ,~
.~, 7~
g requires simpler computational efforts to produce esti-mates of complex conductivity distributions than were required by prior methods.
The above, and other objects and advantages of the present invention will bPcome more apparent from a detailed description of the preferred embodiments when read in conjunction with the drawings.
lQ FIGURE 1 is a diagrammatic representation of the survey field.
The field layout for a survey in accordance with the present invention is shown in Figure 1. A suitable site in the survey area for the magnetic field measurements is selected, preferably rernote from localized non-natural sources of electromagnetic disturbance such as generators, high tension power lines, radio transmitters and busy roads. Magnetic field sensors 10 and 12, of a convention-al type, are installed so as to measure the fluctuationsof the horizontal component of the earth's magnetic field in two non-parallel directions. In the preferred embodiment, these sensors are at right angles to each other, ~ith one sensor parallel to the proposed survey line. Readings from the magnetic field sensors 10 and 12 may be transmitted to a processor station 14 in real time by radio telemetry, as indicated in FIGURE 1, or by direct cable transmission or over telephone lines.
In an aIternative embodiment, the data is recorded in a storage medium such as an electronic computer memory~
magnetic tape or disc resident at the recording site. The stored data could either be transmitted to the processor stations in delayed time or the storage medium containing ~; 35 the data could be phys~ically transported to the processor ~,.
3L2~6~7~
station. This allows independent, unattended operation of the recording stations.
While a single magnetic measurement station is pre-ferred as giving coherent reference signals, if circum-stances demand more than one magnetic station location may be used. If it is necessary as the survey progresses to move the magnetic station, a second station with two sen-sors similarly aligned to the first is set up at the new position, and simultaneous measurements are made for a suitable period in order to translate data obtained using the second station to the same frame of reference as data obtained using the first station, a procedure familiar to those skilled in the art.
A survey line 16 is then determined. This line is preferably straight, but the method may also be used with curved survey lines. Electrodes 18 are installed in elec-trical contact with the earth along the survey line, pref-erably spaced e~uidistant from each other. Adjacent elec-trodes are connected by wires to a potential difference sensor 20. Readings from the potential difference sensors 20 are transmitted to the processor station 14 by radio, cable or telephone or they may be recorded on a mass storage medium at the recording site for subsequent transfer to the processor station, as in the case of the magnetic readings.
The E (electrical) field component is o~tained by dividing the measured potential difference between two electrodes by the electrode spacing L. This type of E
~ield measurement is in ~act a weighted average of the E
~ield component along the wire connecting the two elec-trodes. The arrangement therefore acts as a spatial low pass~filter in wave number space, with an inherent weight-ng factor of a rectangle or boxcar function. The ...
response of the sensor in wave number space is proportion-al to a sine (sinx/x) function with a cutoff wave number somewhat less than the reciprocal of 1,. Those skilled in the art will therefore be able to select a suitable value of L so as to provide the desired survey rasolution.
A se~uence of component E field measurements is made along the survey line 16. As shown in Figure 1, the sen-sors are installed end to end. This arrangement spatially low pass filters the continuous E field component along the path of the survey, then samples this filtered compon-ent at the regular spatial interval L. This type of -sampling is consistent with the requirements of the well-known Sampling Theorem necessary to prevent aliasing of the sampled data. The length of the spread of measure-ments on either side of the points for which it is desired to obtain conductivity data should be approximately equal to the maximum depth to be investigated. The E field measurements may be made simultaneously over the whole survey line or they may be made at one time over one group of adjacent electrodes and then at a later time over another group until the whole survey line has been covPred in se~uence. This latter method is advantageous when surveying to great depth so that the required spread of measurements is very lon~, and simultaneous measurements over the whole line would therefore be impracticable.
When sequential measurements are made, a small overlap in the membership of adjacent groups may be used to provide redundancy to assist in the reduction of noise, but this slows down the survey so preferably it is only used when noise is a serious problem.
When the survey line 16 is curved the process must be modified. First, a low pass filter in the form of a 35 ~weighted sum is~applied to the impedances describing the relationships between the components of the electric field .
i;73 - ~2 -tangential to each of a group of adjacent path se~ments and the component of the magnetic field perpendicular to an effective direction defined for th~ group of segments.
The simple cosine bell weighting function is also used for a curved survey line.
The effective direction of the line segments at a point ~ locating the center of the seg~lent group is deter-mined by defining the angle eo(~) between the effective direction of the group and an arbitrary reference direction, say north. The defining relationship for the angle is N2 (CY) ~ Wi(~)SIN~i i=Nl (CY) eot~) = tan~~ _ N2 (~) ~ Wi (~) COS~i i=N1(~) where the Wi(~) are the cosine bell weights, N1(~) and N2(~) are the lower and upper limits for the sum required to include all of khe non-zero weights over the length of the cosine bell and the ~i are the angles between the survey line se~ments and the reference direction.
The sum of the segment impedances weighted by the cosine bell is then used to compute an apparent depth of penetration. Also, an effective length~ LE(~)l for the survPy line segments in the vicinity of the point a is computed from the defining relation N2 (~) I, ~ Wi(~)COS~o(~
LE(~)= W~ax i=N
`~' -~LZ~ 7~
where L is the length of the individual path segments, all of which are assumed to be equal in length, and WmaX is the largest one of the Wi weights included in the sum.
For the case of thP curved survey path the effective length, LE(~), rather than the length of the cosine bell fun~tion, is required to be a specified multiple of the apparent depth of penetration.
The required value o~ LE(~) is obtained by a process o~ iteration. The process begins by making an initial guess of a suitable length for the cosine bell function.
The weights, Wi(~) are then computed and used to obtain the angle ~O(~) and the effective length LE(~. The apparent depth of penetration is computed from the weighted sum o~ the segment impedances. The effective length and the depth of penetration ara comparecl and the dif~erence between the computed and specified multiple used to determine a new length for the cosine bell. ~he process is repeated in iterative fashion until a suitable measure of convergence is achieved. The procedure is used at each frequency and position on the survey line 16, to be examined.
In the preferred embodiment impedances which are the ratio o~ the fre~uency components of the E field measurement to those o~ the magnetic field mPasured simultaneously in the orthogonal direction to the E ~ield ara then calculated. This impedance designated Zxy is an element of a rank two impedance tensor relationship between the spatial components of the horizontal el~ctric and magnetiç fields. Zxy is defined by the equation Ex = ZxxHx ~ Zx~H~
where Hx is the transform of the magnetic field measure~
ment parallel to the E field measurement, an~ Hy is the transform of the magnetic field mea~uremen~ orthogQnal to the ~ ~ield. The impedance Zxy may be estimated from -14~ 73 the recorded data and the defining equation by any one of a variety of techniques familiar to those skilled i.n the prior art of magnetotelluriçs.
In an alternative embodiment of this invention, the electrical field may also be measured in the direction at right angles to the survey line. In such a case another element of the impedance tensor designated Zxy and defined by the equation Ey = ZyxHx + ZyyHy may also be estimated from the frequency components of the measured data.
It has been discovered as part of this invention that the function Zxy in three dimensional analysis behaves similarly to the E parallel impedance function in conven-tional two dimensional magnetotelluric analysis, except at large wave number values. In this invention the desirable qualities of the E parallel function are obtained in the case of three dimensional conductivlty variation in the earth by attenuating the dependence of the electric field as reflected in Zxy at large wave number values by low pass filtering with respect to wave number space.
It has further bee.n discovered that a critical factor of the low pass filtering operation is the value of the cutoff wave number. It has bee.n determined that, with respect to increasing values of wave number, Zxy and the 2-D E parallel function begin to differ appreciably at the value of the wave length equal to the depth of penetration ~; ~ of an electromagnetic wave into the earth, therefore the reciprocal of this value is an appropriate cutoff, although other values may be used within the scope of the inventi OIl .
, :
, It is well known in the art that, in the wave number and space domains, the linear operation that corresponds to low pass Eiltering is convolution with a kernel weight-ing function. The spatial length of the kernel function is reciprocal with the cutoff wave number so that the length of the kernel increases with decreasing wave number. In order to perform the weighting operation adequately at all wave numbers, it is therefore necessary to have enough spread of E field measuremsnts at each point at which a conductivity sounding is to be obtained so that the longest kernel function required to spatially low pass filter the Zxy impedances can be accommodated.
In the preferred embodiment, the simple cosine bell function may be used as the weighting kernel function.
The exact length of the cosine bell function to be used at any one frequency at any position along the survey line 16 is determined in the preferred embodiment by an iterative method. At each frequency and position to be examined an initial guess of a suitable length of the cosine bell function is made. The impedances are then weighted by that function and the sum of the impedances used to compute an apparent depth of penetration. The length of the cosine function is required to be a specified multiple of this depth of penetration. The preferred embodiment uses a multiple equal to one. The length of the cosine bell is compared to the depth o:E
penetration and the diEference between the computed and specified multiple used to determine a new length for the cosine bell. This process is repeated until a suitable measure of convergence of the computed to the speciied multiple is obtained. In practice this iterative procedure has been found to work well and converge rapidly.
, .
:' -16~ 73 When all the E field measurements over the entire survey path are made simultaneously, an alternative method of obtaining weighted impedances may be used. The weight~
ing procedure, ayain using a cosine bell function of suit-able lenyth, is performed on the E field measurementstransformed to the fre~uenc y domain, and those weighted measurements are then used in the impedance tensor to obtain weighted Zx~ values.
Once the impedances have been spatially weighted at all frequencies of interest the functions are inverted to obtain estimates of conductivity versus depth along the survey path. The inversion procedures used in the preferred embodiment are those used in the prior art one dimension analysis. It would be possible to use more complicated procedures in cases where the increase in accuracy obtained by the use of such procedures justifies the considerable increase in computational effort required by their use.
The method by which the various computations are made is not part of the present invention, but it will be obvious to those skilled in the art that such computations may be conveniently made ~y use of a programmed digital electronic computer. Programs for performing the Fourier transforms, the cosine bell weighting function and the one d1mension inversion procedure are all readily available.
The estimated conductivity at any point in a survey made according to the method of the present invention is a weighted a~erage of the subsurface conductivities both directly underneath, and lateral of, the survey line.
When the bulk of a conductive feature lies lateral of the survey line but is still evident in the cross saction the section is said to be "side-swiped". According to pub-lished theoretical studies, the vertical component of the -17~ 67~
magnetic field at the point of the survey may be used to diagnose a side swipe. Therefore, in an alternative embodiment of this invention, the vertical component of the ma~netic field is measured at the points along the survey path, and these measurements used according to the published methods of analysis to determine whether the cross sections determined by the method described above contain side swipes.
In situations where unusual inhomogeneity of the magnetic field is encountered or anticipated, or where the environment is unusually noisy, or where a higher degree of accuracy is required, additional measurements of the magnetic field in two non-parallel horizontal directions can be made. These measurements may be made along the survey line, but more preferably at points througnout the survey area. The two directions are preferably orthogonal, with one direction preferably being parallel to the survey line.
If the x and y components of the magnetic field are measured at N separate locations, either simultaneously or at different times, the impedances Zax and Zay to be estimated for each span of the electric field array, a span being the distance between two adjacent electrodes, are assumed to satisfy the expression Ea - Zax ( ~ Ui Hxi) + Zay ~ Vi Hyl) where Ui and Vi are appropria~e weighting factors which may be determined by Yarious means familiar to those skilled in the art.
Designating the x and y field components at the reference site Rx and Ry, these field components being electric, magnetic or a combination of the two, the . !.11 -18~ 3 relationship to the x and y magnetic components at the ith site, Hxi and Hyi, is given by Hxi = Txxi Rx + Txyi Ry Hyi = Tyxi Rx ~ Tyyi Ry The transfer ratios, Txii, Txyi, Tyxi and Tyyi may be determined by any of the known referencing techniques conventionally used in magnetotelluric surveying.
Substituting these expression into the equation for Ea gives Ea = Zax ~J (Vi Txwi ~ Vi Tyxi) Rx + Zay ~Vi Txyi + Vi Tyyi) Ry From this Zax and 2ay may be obtained from the measured values of Ea, Rx and Ry by conventional means. The impedances thus determined for each span of the electric field array are then filtered by a process equivalent to convolution with a kernel weighting function as described above. The earth conductivity distributions calculated in this way should be more accurate than those involving single site magnetic measurements, although the additional magnetic measurements would make the survey more expensive to perform.
The embodiments of the invention described ahove relate to survey methods using alternating currents, whether naturally or artificially induced. The invention may also be applied to electrical surveying using a direct current, controlled source of excitation. In this type of surveying, the apparent depth of penetration into the earth of the signals is related to the distance be~ween the source and the point at which the electric field is measured. An electric field is excited in the earth by a :
:
:, -19~ 3 direct current controlled source, techniques for which are well known to those skilled in the art. The excited electric field is measured along a survey line at a distance from the source by the technique described above, namely measuring the potential diference between each pair of adjacent electrodes in the array, and dividing the measured potential difference by the spacing between the electrodes. The measured electric field data is then converted to apparent resistances, computed as the ratio of the measured electric field to the current put out by the controlled source, using conventional processing techniques. Weighted averages of the apparent r~sistances for predetermined distances from the source are obtained in such a way that the process is e~uivalent to using the low pass filtering technique utilizing a kernel weighting function on the electric field along the survey path, as described above. In this case the spatial length o~ the kernel function is reciprocal with a cutoff wave number which is inversely related to the distance between the source and the electric field measurement array.
It is an advantage of the present invention that it permits use of a much simpler configuration o electric field sensors than was possible with the prior magnetotel-luric survey methods. This will be of particular benefitin offshore surveying, and in surveying in very difficult terrain. Another advantage, which will be of benefit in the same situations, is that the magnetic field may be measured in a much simpler manner. Only one magnetic sensor site is required, which in offshore surveying could be on shore, for surveys near the coastline, or positioned adjacent to a fixed offshore platform. Further, unlike conventional magnetotelluric methods, measurement of the ; vertical component of the magnetic field is not required ~to make the suFvey although it may optionally be made to .~ .
, ' ' . .
~ 73 add information relating to structures lateral of the survey line.
It is a further advantage of $his invention that it can be readily adapted to a wide range of survey applications. Without in any way limiting the scope of the invention herein claimed and disclosed, it may be used in prospecting for minerals and hydrocarbons, in locating underground pipes and utility conduits, in detecting seepage of radioactive or toxic materials from storage sites, in detecting underground water supplies, in archeological research and in probing the conductivity of tissue, bones and organs within the body.
Additional advantages and modifications will be readily apparent to those skilled in the art. The inven-tion in its broader aspects is therafore not limited to the specific details, representative apparatus or the illustrative example shown and described. Accordingly, departures may be made from the detail without departing from the spirit or scope of the disclosed general inven-; tive concept.
:: `
:
1~
.
Claims (33)
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A method of obtaining measurements of variations in the earth's magnetic and electric fields for use in an electromagnetic geophysical survey, comprising:
measuring the variations in the earth's magnetic field in at least two horizontal non-parallel directions at a fixed reference point in a survey area; and simultaneously with the measurement of the variations in the magnetic field, measuring variations in the earth's electric field in a direction parallel to a survey line in the survey area at a plurality of points along the survey line spaced so as to sufficiently sample spatial variations of the electric field along the survey line.
measuring the variations in the earth's magnetic field in at least two horizontal non-parallel directions at a fixed reference point in a survey area; and simultaneously with the measurement of the variations in the magnetic field, measuring variations in the earth's electric field in a direction parallel to a survey line in the survey area at a plurality of points along the survey line spaced so as to sufficiently sample spatial variations of the electric field along the survey line.
2. A method as described in claim 1, wherein the two directions in which the earth's magnetic field is measured are orthogonal to one another.
3. A method as described in claim 2, wherein one of the directions in which the variations in the earth's magnetic field is measured is substantially parallel to the survey line.
4. A method as described in claim 1, wherein the step of measuring the variations in the earth's electric field parallel to the survey line consists of measuring the variations in the potential differences between one or more pairs of adjacent electrodes in electrical contact with the earth and dividing the measured potential difference between each pair of adjacent electrodes by the distance between that pair.
5. A method as described in claim 4, wherein the electrodes are spaced at substantially equal intervals along the survey line.
6. A method as described in claims 1 or 4, wherein the survey line is substantially straight.
7. A method as described in claim 4, wherein the variations in the potential differences between each pair of adjacent electrodes are measured simultaneously.
8. A method as described in claim 4, wherein the variations in the potential differences between each pair of adjacent electrodes in a group of adjacent electrodes consisting of less than all the electrodes are measured simultaneously and the measurements for each of the groups are made sequentially.
9. A method as described in claim 8, wherein at least one of the electrode pairs in one group is also a member of the adjacent group.
10. A method of obtaining measurements of variations in the earth's magnetic and electric fields for use in an electromagnetic geophysical survey, comprising:
measuring the variations in the earths magnetic field in at least two horizontal non-parallel directions at a fixed reference point in the survey area; and simultaneously with the measurement of the variations in the magnetic field, measuring the variations in the earth's electric field parallel to and orthogonal to a survey line in the survey area at a plurality of points spaced so as to sufficiently sample spatial variations of measured components of the electric field along the survey line.
measuring the variations in the earths magnetic field in at least two horizontal non-parallel directions at a fixed reference point in the survey area; and simultaneously with the measurement of the variations in the magnetic field, measuring the variations in the earth's electric field parallel to and orthogonal to a survey line in the survey area at a plurality of points spaced so as to sufficiently sample spatial variations of measured components of the electric field along the survey line.
11. A method as described in claim 10, wherein the two directions in which the earth's magnetic field is measured are orthogonal to one another.
12. A method as described in claim 10, wherein one of the directions in which the variations in the earth's magnetic field is measured is substantially parallel to the survey line.
13. A method as described in claim 10, wherein the points at which measurements of the earth's electric field are made are spaced at substantially equal intervals along the survey line.
14. A method as described in claim 10, wherein the survey line is substantially straight.
15. A method as described in claim 10, wherein the variations in the earth's electrical field at each point along the survey line are measured simultaneously.
16. A method as described in claim 10, wherein the variations in the earth's electric field at a group of adjacent points along the survey line consisting of less than all the points along the survey line are measured simultaneously and the measurements for each of the groups are made sequentially.
17. A method as described in claim 16, wherein at least one of the points in one group is also a member of the adjacent group.
18. A method as described in claim 10, wherein the step of measuring the variations in the earth's electric field parallel to the survey line consists of measuring the variations in the potential differences between one or more pairs of adjacent electrodes in electrical contact with the earth and dividing the measured potential difference between each pair of adjacent electrodes by the distance between that pair.
19. A method as described in claim 10, wherein the step of measuring the earth's electric field orthogonal to the survey line consists of measuring the variations in the potential difference between a pair of electrodes in electrical contact with the earth, situated on a line orthogonal to the survey line, and dividing the measured potential difference by the distance between the electrodes.
20. A method as described in claim 1 or claim 10, further comprising the step of simultaneously measuring the variations in the earth's magnetic field in a vertical direction at a plurality of points along the survey line.
21. In a method of obtaining measurements of variations in the earth's magnetic and electric fields for use in an electromagnetic geophysical survey method which comprises measuring the variations in the earth's magnetic field in at least two horizontal non-parallel directions at a first fixed reference point in the survey area and simultaneously measuring variations in the earth's electric field in at least a first direction, which first direction is parallel to a survey line in the survey area, at a plurality of points spaced so as to sufficiently sample spatial variations of the measured components of the electric field along the survey line, a method of establishing a second fixed reference point in the survey area for the measurement of variations in the earth's magnetic field, comprising:
measuring the earth's magnetic field at the second reference point in at least two horizontal directions substantially parallel to the horizontal directions in which the variations in the earth's magnetic field are measured at the first reference point; and simultaneously continuing to measure the variations in the earth's magnetic field at the first reference point for a period sufficient to permit data obtained from the second reference point to be correlated to data obtained from the first reference point.
measuring the earth's magnetic field at the second reference point in at least two horizontal directions substantially parallel to the horizontal directions in which the variations in the earth's magnetic field are measured at the first reference point; and simultaneously continuing to measure the variations in the earth's magnetic field at the first reference point for a period sufficient to permit data obtained from the second reference point to be correlated to data obtained from the first reference point.
22. In an electromagnetic geophysical survey method in which the variations in the earth's magnetic field in at least two horizontal directions are measured at at least one reference point in the survey area and the variations in the earth's electric field parallel to a survey line in the survey area are measured at a plurality of survey points spaced so as to sufficiently sample the electric field along the survey line simultaneously with the measurements of the variations in the magnetic field, a method of deriving the conductivity of the earth below the survey line, comprising:
transforming the measured variations in the magnetic field and electrical field into frequency components;
calculating as a function of frequency the horizontal component of the earth's magnetic field orthogonal to the direction of the measured electric field at each of the survey points from the measurements of the magnetic field in the two non-parallel directions;
calculating as a function of frequency the impedance at each survey point, the impedance being the ratio between the measured electric field at that point and the horizontal component of the earth's magnetic field orthogonal to the direction of the measured electrical field;
calculating by a weighting process equivalent to applying a low pass filter to the electric field in wavenumber space with a cutoff wavenumber for the filter that varies substantially inversely proportional to the effective depth of penetration into the earth of an electromagnetic wave of that frequency for predetermined frequencies the weighted averages of the impedances such that the number of impedances entered into each weighted average increases with decreasing frequency; and calculating the distribution of conductivity in the earth below the survey line as a function of depth from the weighted averages of the impedances.
transforming the measured variations in the magnetic field and electrical field into frequency components;
calculating as a function of frequency the horizontal component of the earth's magnetic field orthogonal to the direction of the measured electric field at each of the survey points from the measurements of the magnetic field in the two non-parallel directions;
calculating as a function of frequency the impedance at each survey point, the impedance being the ratio between the measured electric field at that point and the horizontal component of the earth's magnetic field orthogonal to the direction of the measured electrical field;
calculating by a weighting process equivalent to applying a low pass filter to the electric field in wavenumber space with a cutoff wavenumber for the filter that varies substantially inversely proportional to the effective depth of penetration into the earth of an electromagnetic wave of that frequency for predetermined frequencies the weighted averages of the impedances such that the number of impedances entered into each weighted average increases with decreasing frequency; and calculating the distribution of conductivity in the earth below the survey line as a function of depth from the weighted averages of the impedances.
23. A method of deriving the conductivity of the earth as described in claim 22, wherein the weighted averages of the impedances are calculated by a weighting process equivalent to convolving the electric field along the survey path with a zero phase, finite length weight function, the width of the weight function for each frequency being determined by selecting an appropriate width, obtaining a weighted average impedance using the selected function, using the weighted average so obtained to calculate an apparent depth of penetration, comparing the calculated depth of penetration to the expected depth of penetration, using the difference between the calculated and expected depths to assist in the selection of a more appropriate width, and repeating the process iteratively until a predetermined accuracy of calculation of the depth of penetration is obtained.
24. In an electromagnetic geophysical survey method in which the variations in the earth's magnetic field in at least two horizontal directions are measured at at least one reference point in the survey area and the variations in the earth's electric field parallel to a survey line are simultaneously measured at each of a plurality of survey points spaced so as to sufficiently sample spatial variations of the electric field along the survey line simultaneously with the measurements of the variations in the magnetic field, a method of deriving the conductivity of the earth below the survey line, comprising:
transforming the measured variations in the potential differences to a function of frequency;
calculating by a weighting process equivalent to convolving the electric field along the survey path with a zero phase finite length weight function the width of which is substantially proportional to the effective depth of penetration into the earth of an electromagnetic wave of that frequency for predetermined frequencies the weighted averages of the measured potential differences such that the number of potential differences entered into each weighted average increases with decreasing frequency;
calculating the horizontal component of the earth's magnetic field orthogonal to the direction of the survey line from the measurements of the magnetic field in the two non-parallel directions;
calculating weighted impedances along the survey line from the ratio of the weighted averages of the potential differences and the horizontal orthogonal component of the magnetic field; and calculating the distribution of conductivity of the earth below the survey line as a function of depth from the weighted impedances.
transforming the measured variations in the potential differences to a function of frequency;
calculating by a weighting process equivalent to convolving the electric field along the survey path with a zero phase finite length weight function the width of which is substantially proportional to the effective depth of penetration into the earth of an electromagnetic wave of that frequency for predetermined frequencies the weighted averages of the measured potential differences such that the number of potential differences entered into each weighted average increases with decreasing frequency;
calculating the horizontal component of the earth's magnetic field orthogonal to the direction of the survey line from the measurements of the magnetic field in the two non-parallel directions;
calculating weighted impedances along the survey line from the ratio of the weighted averages of the potential differences and the horizontal orthogonal component of the magnetic field; and calculating the distribution of conductivity of the earth below the survey line as a function of depth from the weighted impedances.
25. A method of obtaining a resistivity survey of the earth's surface, comprising:
providing a direct current controlled source of electrical excitation of the earth's surface;
simultaneously measuring the electric field excited in the earth's surface by the direct current controlled source at a plurality of points spaced so as to sufficiently sample variations of the electric field along a survey line at a distance from the controlled source;
calculating the apparent resistances at each of the points along the survey line from the ratio of the measured electric field to the source current;
calculating by a weighting process equivalent to applying a low pass filter to the electric field in wavenumber space with a cutoff wavenumber for the filter that varies substantially inversely proportional to the effective depth of penetration into the earth of the direct current for that distance for predetermined distances from the source the weighted averages of the apparent resistances such that the number of apparent resistances entered into each weighted average increases with increasing distance; and calculating the distribution of resistivity of the earth below the survey line as a function of depth from the weighted apparent resistances.
providing a direct current controlled source of electrical excitation of the earth's surface;
simultaneously measuring the electric field excited in the earth's surface by the direct current controlled source at a plurality of points spaced so as to sufficiently sample variations of the electric field along a survey line at a distance from the controlled source;
calculating the apparent resistances at each of the points along the survey line from the ratio of the measured electric field to the source current;
calculating by a weighting process equivalent to applying a low pass filter to the electric field in wavenumber space with a cutoff wavenumber for the filter that varies substantially inversely proportional to the effective depth of penetration into the earth of the direct current for that distance for predetermined distances from the source the weighted averages of the apparent resistances such that the number of apparent resistances entered into each weighted average increases with increasing distance; and calculating the distribution of resistivity of the earth below the survey line as a function of depth from the weighted apparent resistances.
26. An electrical geophysical survey method, comprising:
providing a direct current controlled source of electrical excitation of the earth's surface; and simultaneously measuring the electric field excited in the earth surface by the direct current controlled source at a plurality of points with spacing sufficiently close to adequately sample all variations of the electric field along a survey line.
providing a direct current controlled source of electrical excitation of the earth's surface; and simultaneously measuring the electric field excited in the earth surface by the direct current controlled source at a plurality of points with spacing sufficiently close to adequately sample all variations of the electric field along a survey line.
27. An electrical geophysical survey method as described in claim 26, wherein the step of measuring the excited electric field along a survey line consists of measuring the variations in the potential differences between one or more pairs of adjacent electrodes in electrical contact with the earth and dividing the measured potential difference between each pair of adjacent electrodes by the distance between that pair.
28. An electrical geophysical survey method as described in claim 27, wherein the electrodes are spaced at substantially equal intervals along the survey line.
29. An electrical geophysical survey method as described in claim 26, wherein the survey line is substantially straight.
30. An electrical geophysical survey method as described in claim 27, wherein the variations in the potential differences between each pair of adjacent electrodes are measured simultaneously.
31. An electrical geophysical survey method as described in claim 27, wherein the variations in the potential differences between each pair of adjacent electrodes in a group of adjacent electrodes consisting of less than all the electrodes are measured simultaneously and the measurements for each of the groups are made sequentially.
32. An electrical geophysical survey method as described in claim 31, wherein at least one of the electrode pairs in one group is also a member of the adjacent group.
33. In an electrical geophysical survey method in which an electric field excited in the earth's surface by a direct current controlled source is measured at a plurality of points spaced to adequately sample the electric field along a survey line at a distance from the controlled source, a method of deriving the resistivity of the earth below the survey line, comprising:
calculating the apparent resistances at each of the points along the survey line from the ratio of the measured electric field to the source current;
calculating by a weighting process equivalent to applying a low pass filter to the electric field in wavenumber space with a cutoff wavenumber for the filter that varies substantially inversely proportional to the effective depth of penetration into the earth of the direct current for that distance for predetermined distances from the source the weighted averages of the apparent resistances such that the number of apparent resistances entered into each weighted average increases with increasing distance; and calculating the distribution of resistivity of the earth below the survey line as a function of depth from the weighted apparent resistance.
calculating the apparent resistances at each of the points along the survey line from the ratio of the measured electric field to the source current;
calculating by a weighting process equivalent to applying a low pass filter to the electric field in wavenumber space with a cutoff wavenumber for the filter that varies substantially inversely proportional to the effective depth of penetration into the earth of the direct current for that distance for predetermined distances from the source the weighted averages of the apparent resistances such that the number of apparent resistances entered into each weighted average increases with increasing distance; and calculating the distribution of resistivity of the earth below the survey line as a function of depth from the weighted apparent resistance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA000508778A CA1246673A (en) | 1986-05-09 | 1986-05-09 | Electromagnetic array profiling survey method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA000508778A CA1246673A (en) | 1986-05-09 | 1986-05-09 | Electromagnetic array profiling survey method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1246673A true CA1246673A (en) | 1988-12-13 |
Family
ID=4133107
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA000508778A Expired CA1246673A (en) | 1986-05-09 | 1986-05-09 | Electromagnetic array profiling survey method |
Country Status (1)
| Country | Link |
|---|---|
| CA (1) | CA1246673A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113031087A (en) * | 2021-03-03 | 2021-06-25 | 王佳馨 | Cross-street opposite-penetration resistivity measurement system and data acquisition method |
-
1986
- 1986-05-09 CA CA000508778A patent/CA1246673A/en not_active Expired
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113031087A (en) * | 2021-03-03 | 2021-06-25 | 王佳馨 | Cross-street opposite-penetration resistivity measurement system and data acquisition method |
| CN113031087B (en) * | 2021-03-03 | 2022-10-28 | 王佳馨 | Cross-street opposite-penetration resistivity measurement system and data acquisition method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US4757262A (en) | Method for geophysical exploration using electromagnetic array | |
| US6603313B1 (en) | Remote reservoir resistivity mapping | |
| Chave et al. | Electrical exploration methods for the seafloor | |
| Spies et al. | Electromagnetic sounding | |
| EP2052267B1 (en) | Time lapse analysis with electromagnetic data | |
| US9625600B2 (en) | Systems and methods for removal of swell noise in marine electromagnetic surveys | |
| US4835473A (en) | Method using a d.c. source for determining resistivity distribution of the earth as a function of depth | |
| USRE40321E1 (en) | Remote reservoir resistivity mapping | |
| Edwards et al. | First results of the MOSES experiment: sea sediment conductivity and thickness determination, Bute Inlet, British Columbia, by magnetometric offshore electrical sounding | |
| US4591791A (en) | Electromagnetic array profiling survey method | |
| US4792761A (en) | Geophysical prospecting with collimated magnetotelluric fields | |
| Edwards et al. | The magnetometric resistivity method | |
| EA011273B1 (en) | System and method for time-distance characteristics in acquisition, processing and imaging of t-csem data exploiting controlled sources in time domain | |
| Saey et al. | Electrical conductivity depth modelling with a multireceiver EMI sensor for prospecting archaeological features | |
| EA022910B1 (en) | Method for conducting electromagnetic survey | |
| US20110320125A1 (en) | Method for marine geoelectrical exploration with electrical current focusing | |
| Manzella | Geophysical methods in geothermal exploration | |
| Nekut | Electromagnetic ray-trace tomography | |
| EP2230534A1 (en) | Method for determining resistivity anistropy from earth electromagnetic responses | |
| AU2014201354B2 (en) | Systems and methods for measuring water properties in electromagnetic marine surveys | |
| Turberg et al. | Joint application of radio-magnetotelluric and electrical imaging surveys in complex subsurface environments | |
| CA1246673A (en) | Electromagnetic array profiling survey method | |
| BR102012030628B1 (en) | METHOD AND APPARATUS TO DETERMINE CHARACTERISTICS OF A TARGET REGION BELOW A BODY OF WATER | |
| Schaa | Rapid approximate 3D inversion of transient electromagnetic data | |
| Ferguson et al. | Electromagnetic mode conversion by surface-conductivity anomalies: applications for conductivity soundings |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKEX | Expiry |