3-d controlled source electromagnetic (csem) interferometry_nikhil prakash

8/8/2019 3-D Controlled Source Electromagnetic (CSEM) Interferometry_Nikhil Prakash

1/14

Yuanzhong Fan1, Roel Snieder1 and Johannes Singer21Center for Wave Phenomena, Department of

Geophysics, Colorado School of Mines2Shell International Exploration and Production,

Houston

ByNikhil Prakash

071809, 4th year GPT


2/14

Controlled Source Electromagnetic (CSEM) is animportant technique in hydrocarbon explorationbecause it uses the large contrast in electrical

resistivity to distinguish between water andhydrocarbons

In a shallow sea environment, the airwave that isrefracted from the air-water interface dominates

the recorded signal at large offsets. herefore, thehydrocarbon detection ability of the CSEM iswea ened because the airwave is independent ofthe properties of the subsurface.


3/14

For a layered earth model, we apply multi-dimensional deconvolution interferometry tosynthetic 3D CSEM data and estimate the

reflection response of the subsurface. hedifference in the models with and without aresistive layer is significantly increased by theemployed interferometric analysis. However,

the required receiver spacing is much denserthan that of current CSEM surveys.


4/14

Snieder (2 6) shows that interferometry can be applied notonly to wavefields, but also to diffusive fields. However,crosscorrelation- based diffusion interferometry requires a volumesource distribution in order to apply this technique successfully (Fanand Snieder, 2 ). he required source distribution is often not

realized in practice; his ma es it impractical to apply cross-correlation-based diffusion interferometry to real applications.

When the receivers are located in a plane and sources areplaced above this plane, a multi-dimensional-deconvolutionapproach is shown to be applicable to a diffusion field(Amundsen et al., 2 6; Slob et al., 2 ; Wapenaar et al., 2 ).


5/14

It also holds for any field which can be decomposed intoupgoing and downgoing components. A generic geometryfor this approach is s etched in Figure . he source isdenoted by the star, and receivers are located on plane 1.is a boundary above the sources, which may or may not bepresent.

where U(xA;xs;w) represents the upgoingfield received at location xA in thefrequency domain due to the source at xs.

he downgoing field is noted by D, and

R(xA;x;w) is the reflection response thatrelates the downgoing field at x to theupgoing field at xA


6/14

The inversion of R from equation (1) is ill-posed because no unique R canbe obtained from a downgoing field D and upgoing field U excited by asingle source. If a source at another position xs is used, a different pair of U

and D is obtained fromthe decomposition.The medium response R, however, remains the same because it isindependent of the source position. This means that the more sources weuse, the more constraints there are on the inversion of R.Therefore, a band-limited medium response R can be accurately inverted

from a band-limited input signal, if a sufficient number of real sources areused.


7/14

If we can successfully apply the multi-dimension-deconvolutioninterferometry (as described in the last section) to CSEM, oneof the receivers is converted into a source and the overburdenis extended upwards to a homogeneous half space (Wapenaaret al., 2 ). Consequently, the air-water interface and

the sea floor are removed, and there is no secondary field refractedfrom the medium above the receivers.

the secondary field is generated only by the subsurface

and therefore it is easier and more accurate to detect theproperties of a target layer.


8/14

A 2D receiver array is used in the syntheticexample because both wavenumbers in the x andy directions are required in the decompositioninto upgoing and downgoing fields, and thesurface integral in equation (1) must be replaced

by the summation of receivers.

The uniformly sampled 2D receiver array is on thesea floor from the position (-1 m, -1 m) to (-1

m, 1 m) with a receiver separation dr of 5 min both x and y directions. The EM source is a

dipole in the x direction with a length of 1 m,and a AC current of 1 A, and an operatingfrequency of .25 Hz.


9/14

Without target

With target

The inline electric field. Due to more cost, weuse only profiles where also there is very less

difference between the plots for with andwithout target layer.


10/14

The implementation of the up-down decomposition follows the theory in theappendix of Wapenaar et al.(2 ). ote that the up-down decomposed field isthe square root of energy flux, not the E (electric) or H (magnetic) field.

where P is the decomposed upgoing and downgoing potential,normalized to energy flux, Q contains the input horizontalE and H fields, and L is the conversion operator.

In our synthetic example, the field is 3D; hence Q containsfour components Ex;Ey;Hx;Hy, and P has four components aswell [P1d ;P1u ;P2d ;P2u ].

ecause the receivers are located at the boundary of the water and the sea floor, wecan choose the parameters for L1 from the upper medium (water) or the lowermedium (sea floor) in the process of the field decomposition. Using the waterparameters for the up-down decomposition, we obtain the upgoing and downgoingfields in the water just above the sea bottom. If the sea floor parameters are used, we

obtain the upgoing and downgoing fields in the sea bottom just below the acquisitionsurface.


11/14

This is a multi-dimensionaldeconvolution problem to compute Rfrom U and D.

In the wave number domain,we can present spatialconvolution in equation (3) bymultiplication

The difference of the impulse response is significant between thetwo models with and without target. Comparing the inline profile ofthe total Ex field (figure 5) with the inline profile of the impulseresponse (figure ) gives much more pronounced differencebetween the models with and without the target.


12/14


13/14

The 3-D synthetic example in this paper shows that the virtual source technique inCSEM can significantly increase the sensitivity of detecting the high-resistivitylayer (such as hydrocarbon reservoir) in the submarine environment.

The required receiver separation dr must be less than the height hs of the

dipole source above the sea bottom to adequately carry on the up-downdecomposition(dr < hs).

This sampling criterion, however, is not practical for two reasons.The first is that in current CSEM surveys, the separationof the receivers is much larger than this (2 times) in the fieldsurvey.

The second is that there is only a line of receivers (insteadof the 2D array we employed) used in current practicalcases.In order to ma e this technique practical, the requirementof the dense 2D receiver networ must be relaxed.Thisis the topic of ongoing research.


14/14

3-d controlled source electromagnetic (csem) interferometry_nikhil prakash

Documents