WO2009032996A2

WO2009032996A2 - Seismic resonance imaging

Info

Publication number: WO2009032996A2
Application number: PCT/US2008/075362
Authority: WO
Inventors: Valeri A. Korneev
Original assignee: The Regents Of The University Of California
Priority date: 2007-09-06
Filing date: 2008-09-05
Publication date: 2009-03-12
Also published as: WO2009032996A3

Abstract

A method is described for first locating and then determining the position of buried objects such as pipes, fluid filled cracks, UXO, tunnels, and the like. By this method, seismic responses are analyzed for the presence of late arriving secondary waves generated by resonant objects in response to a seismic wave shot. Upon detection of such resonant waves, indicating the presence of a buried object, the location of the object can be found as providing a best fit of a computed field and recorded data. A buried resonant object can be excited remotely by illumination of a distant active source.

Description

SEISMIC RESONANCE IMAGING

Inventor: Valeri Korneev

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit of priority under 35 USC Section 119(e) to Provisional Application Serial Number 60/970,471, filed September 6, 2007, which application is hereby incorporated by reference in its entirety.

STATEMENT OF GOVERNMENTAL SUPPORT

[0002] The invention described and claimed herein was made in part utilizing funds supplied by the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. The U.S. Government has certain rights in this invention.

FIELD OF THE INVENTION

[0003] The present invention relates to seismic imaging, and more particularly to imaging subsurface objects using resonant seismic emissions.

BACKGROUND OF THE INVENTION

[0004] Imaging of shallow subsurface heterogeneities has a variety of important applications. Those applications include detection and location of different kinds of local inhomogeneities such as tunnels, pipes, buried containers, ground-filled excavations, unexploded ordinances (UXO), fluid-filled fractures, mine shafts, and the like. Being high contrast scatterers, these objects are capable of generating strong scattered waves where primary PP, PS, and SS waves carry away most of the energy which was brought by incident waves. This conclusion follows from numerical and analytical results obtained from canonical solutions for spheres and cylinders.

[0005] For both high- and low- velocity objects the primary scattered waves have the same order of magnitude as incident waves. While low-contrast objects (i.e. objects having an impedance relatively close to that of the embedding elastic medium [where impedance is defined as the product of velocity times density]) effectively radiate most of the energy soon after impact, high-contrast objects (i.e. objects having an impedance significantly different from that of the embedding medium) trap some fraction of incident wave energy in the form of durable circumferential waves, which propagate, rotating along the interface between the object and the embedding medium. The seismic energy from these waves is slowly released, long after the initial impact. Release of this trapped energy occurs primarily as a resonant emission of shear waves and can be detected as sharp resonant peaks in amplitude spectrums of single records. These circumferential waves include surface Rayleigh-type waves (propagating mostly in the embedding medium), Stoneley waves (propagating mostly in the fluid, if present), and Frantz waves (body waves trapped in the object because of its curvature). Strong impedance contrast ensures small radiation loss for circumferential waves, which slowly decay in amplitude while rotating inside and around the object.

[0006] Most of the secondary resonant-scattered energy (reflected from an object's boundaries or circling around the object) radiate in the embedding medium as shear waves. This is reflective of the fact that S- waves are slower than P- waves and that energy conversion is generally larger for slow velocity waves. In practice, the amplitudes ratio for S- to P- waves is increased even more because the local amplification factors are inversely proportional to wave velocities. The possibility of neglecting P- waves in late scattering arrivals simplifies imaging as is demonstrated for the field and modeled data of the Example.

[0007] In practice, there are several classes of objects that respond with velocities very different than in the embedding elastic media (e.g. the ground). These include tunnels, pipes and cavities filled with a fluid (such as a liquid or air). Low gas densities in such structures result in long-living circumferential waves (primarily of the Stoneley type). Generally, water and other underground liquids (such as oil) have comparatively low velocities of about 1500 m/s and being contained in a volume can trap energy. Unexploded ordinance (UXOs) present an example of high- velocity contrast objects.

[0008] Another class of wave-trapping objects are localized low- velocity zones, which can have a natural low velocity (like a part of a coal seam) or which result from some impact such as a filled excavation pit of loosened rock/soil which have smaller elastic modules compared with the embedding medium. Still another class of wave trapping objects includes fluid-filled fractures in rocks. Stoneley slow waves can have very low velocities in such fractures which makes resonances possible even for seismic frequencies (with frequencies less than 100 Hz).

BRIEF DESCRIPTION OF THE INVENTION

[0009] The present invention provides a method for detecting the subsurface presence of such features as have been described above, in situations where the transmission velocities between the feature and the embedding medium are sufficiently different such that the embedded feature generates resonant waves which can be detected. The larger the contrast in impedance, the more readily the object may be detected. By appropriate placement of detectors and monitoring of the resultant resonant wave pulses, both the presence of the subsurface feature, its location below the surface (i.e. its location in three dimensions) and its approximate size may be determined.

[0010] In an embodiment of the invention, a plurality of seismic detectors, such as standard geophones, are placed over an area of surface to be examined for the presence of underground features. The sensors in one embodiment can be disposed along a line when the orientation of the object is known or arrayed as an ordered two to three dimensional grid. In another embodiment they may be randomly dispersed over a generalized area. A seismic source is provided for generating a seismic impulse wave. The source may be a simple hammer and plate combination, an automatic hammer, an explosive charge, a motor vehicle or the like. Natural seismic noise, caused by a variety of possible unknown sources can similarly be used provided they generate over a long enough time for these waves to be recorded. Each sensor is linked either by cable or wirelessly to a monitor/computer for recording/storing and displaying of the processed output of the sensors.

[0011] To determine the presence of a feature, the seismic source is activated, generating an impulse wave which travels in all directions through the ground. The seismic response at each of the sensors is recorded, and the initial response from the high energy direct incident waves ignored (though this information can be used for S- wave velocity (V_s) estimates). This is done by recording the initial response of two sensors of known distance from the source, and calculating the velocity of the S- wave by dividing the difference of those distances by the travel time difference recorded at those sensors.). In one embodiment, the gain of the monitoring device is turned up so that any late arriving resonant wave responses can be detected. Once a resonant wave response is detected (at the same recorded time-stable spectral peak for adjacent sensors over several shot gathers), which signals the presence of a buried feature, the location of the buried feature can then be determined. This is done through a series of shot gathers taken either at predetermined times when using an active controlled source, or at randomly chosen times when using ambient noise waves.

[0012] As used herein, the actuation of a seismic source is known as a shot, and the set of recorded responses at various sensors/receivers is referred to as a shot gather. Simultaneous detection of the noise waves by a set of sensors is also referred as a shot and all the records made after such detection also compose a shot gather. In this context, several shot gathers are achieved, in one embodiment re-positioning the plurality of sensors until, based upon the observed responses, it is determined that several sensors/receivers record the same resonance emission of the buried feature. This determination is made by comparison of the responses of adjacent sensors, in the manner more fully explained below in the Detailed Description and the Example. In this manner the approximate lateral position of the subject feature is established. In another embodiment, without taking additional readings, and using software tools, the theoretical data for each potential feature/object position can be computed (assuming the presence of that feature generates monochromatic waves at the detected resonant frequency) and compared to the field recorded data. The theoretical location that provides the best fit between computed and recorded data indicates the true location of the object.

[0013] Once the lateral position of the feature is approximately determined (by looking at that subset of sensors that revealed the presence of resonant peaks of the same frequency), the accurate location of the buried feature may be readily calculated. This is done, using a computer, by first computing a model of the subsurface system response, based upon the evaluated shear- wave propagation velocity V₈ of the embedding elastic medium, modeling expected sensor responses for different locations (as is later explained in the Detailed Description); and then matching the calculated responses to those measured in the field.

[0014] Using the values of resonant frequencies and prior knowledge about the object's shape and velocity of the filling material (size) it is possible to determine the object's size (velocity of the filling material) using theoretical solutions for the objects of the same shape.

BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flow diagram of various steps utilized according to a method of this invention for identifying the location of embedded feature.

[0016] Figure 2 is a schematic illustration of a field arrangement for the detection of resonant emissions. Controlled sources may be underground and/or at the surface.

[0017] The following Figures relate to the data generated in the Example.

[0018] Figure 3 shows the shot gathers in the Example for field data recorded above the buried barrel, the direct S-waves visible.

[0019] Figure 4 shows the amplitude spectrum of automatic gain controlled late arrivals with a sharp peak at 78 Hz.

[0020] Figure 5 shows the shot gathers after automatic gain control (AGC) and band-pass filtering around 78 Hz, showing hyperbolic signatures of a secondary source.

[0021] Figure 6 shows the image of a buried barrel.

[0022] Figure 7 shows the partial scattering cross-section for a fluid-filled sphere for P- incident waves (a) and for S- incident waves (b).

[0023] Figure 8 shows single shot gather for synthetic field computed for a fluid-filled tunnel model. [0024] Figure 9 is a snap-shot showing propagation of circumferential waves along the object.

[0025] Figure 10 shows amplitude spectrum of automatic gain controlled late arrivals with a sharp peak at 53 Hz.

[0026] Figure 11 shows late arrivals of modeled data after AGC (automatic gain control) and band pass filtering around the resonance frequency, having repeating quasi-hyperbolic patterns similar to one observed for field data (Figure 5).

[0027] Figure 12 shows the migrated image of the modeled object is similar to one obtained from field data (Figure 6).

[0028] Figure 13 shows theoretical resonances for a 1 mm thick, 4 m long fracture filled with water.

[0029] Figure 14 shows theoretical resonances for a 1 mm thick, 4 m long fracture filled with oil.

[0030] Figure 15 shows the partial scattering cross-section for a fluid-filled sphere for S- incident waves for extended frequency band.

DETAILED DESCRIPTION OF THE INVENTION

[0031] The invention will now be described in detail, with particular reference first to Figure 1. In the practice of the methods of this invention, in one embodiment, a seismic source, a plurality of seismic sensors, a recording station and a computer are employed.

[0032] As noted in the Brief Description of the Invention, the intent of a field investigation using the methods described herein is to determine the presence or absence of a buried feature such as a pipe, tunnel or fluid-filled crack within a selected area of examination. And if present, to the extent possible, determine both the location of the feature and its shape. Broadly, this is done by generating a seismic wave in step 104 initiated using the provided seismic source or ambient seismic noise (step 100), and analyzing the information in step 108 detected in step 106 by sensors deployed in step 102. [0033] The seismic source may be literally anything that can generate a seismic wave. In one embodiment, the one of the illustrated example, the source can consist of no more than a metal plate laid upon the ground, and a sledge hammer which when brought into quick contact with the plate, generates a pulse. In another embodiment the source can be a piezeo-electric device which when energized can create a propagating seismic wave, an automatic hammer, pile driver, a low level explosive device, vibrator truck, rifle and the like. The knowledge about the location of the source is not necessary. The nature of the source is not critical.

[0034] Seismic sensors, hereinafter alternatively referred to as receivers, record vibrations in forms of traces U₁ (t), i = 1, 2, ..., / where / is the number of sensors, and t is time in some recording interval T_begιn ≤ t ≤ T_md . In the case of use of an active source, T_{be m} is the source activation time, hi the case of use of an ambient noise source, T_{be m} is an arbitrarily chosen start time of recording. Here, exact knowledge about T_b is not important. Sensors are placed in a set of / locations η = (X^y₁, Z₁) around the volume where the target object might be located. Analyses are performed for the parts of each record U₁ it) = U₁ (t) , for a time (t) interval which does not contain strong direct arrivals, ground roll and primary scattered waves. Records U₁ (Y) may or may not be subjected to Automatic Gain Control (AGC) procedures.

[0035] The employed sensors can be any readily available commercial device, including vertical geophones, capable of sensing both reflected primary waves and the weaker resonant waves generated by the buried resonant object, or feature. These sensors by be physically linked, connected by electric cables to carry their detected responses to the seismic monitor. In the illustrative example, the sensors are provide as a linked chain, each sensor spaced about a meter from its adjacent sensor and its connecting wire bundled with those of the other sensors of the linked line. The number of linked sensors, and thus the length of the line itself are not critical. The longer the line however, the greater the area examined, and accordingly the more information gathered in each shot gather. In another embodiment, the sensors may be placed individually, linked to the seismic monitor by wireless transmitters associated with each sensor. In this embodiment, the sensors may more easily be placed, such as the intersection of multiple non-parallel grid lines or in an arbitrary grid configuration, to one that is random over a given surface area, to thus provide three-dimensional coverage.

[0036] The seismic recording station employed provides simultaneous recording for all sensors. Such recording station devices can be readily obtained commercially, and the exact type or brand of such recording system is not critical to this invention so long as it is capable of performing the functions above described. The computer used to perform the calculations can likewise be a standard item, programmed to perform the calculations required as described herein. The recording station should also be capable of copying data into computer memory, which is the common case.

[0037] In the method of the invention, a number of sensors are laid out over the area to be investigated, connected to the seismic recording station (either, by wire or wirelessly), and the seismic source repeatedly positioned in the field. Alternatively the receivers can be left in the field with a stationary recording system where the "source" may consist of ambient noise, the system deployed for the long term monitoring of an area of interest, where the post deployment detection of stable resonant peaks indicates the appearance of new feature. To investigate a given area using a linked line of sensors, at steps 108 and 110, a multiplicity of shot gathers are conducted, and the responses recorded. The source and its associated line of sensors can be moved incrementally, both longitudinally and transversely. In one embodiment, the source and line of sensors are moved after each single shot gather. In another embodiment of the invention where the sensors are laid out in a grid pattern, the source can be moved from one grid line to the next in the course of the investigation. In yet another embodiment, separate seismic sources can be positioned at an end of each of the grid lines.

[0038] The detection of resonant peaks indicates a presence of an object of potential interest in the vicinity of sensor installation. For detection one can compute the complex spectrums -?,(/) of the records U₁ (Y) using an inverse Fourier transform. Spectrums s_t (/) are the functions of frequency / . Objects are identified by the presence of sharp frequency peaks of amplitude spectra J,(/)| at same frequency f_res (called resonant frequency) detected at a local subset of sensors, meaning that a resonance frequency appears at several adjacent sensors. It is to be noted that one object may exhibit several resonant frequencies. [ As illustrated in Fig 13, 14 and 15, which can be used to assist in the identification of the nature of the object detected.]

[0039] Given the fact that different objects may have their own unique frequency response, it also may be possible to determine the nature of a buried object from an observed response. This can be done by field trials in which various objects are buried in defined soils and the response frequency and its intensity is recorded and stored to create a database. In the field, when resonant wave responses are detected, these responses can then be compared to those of the stored database to permit ready identification of the nature of the buried feature. A knowledge about an object's shape (e.g. tunnel) and filling material (air) enables one to also determine its effective diameter using available theoretical solutions where the theoretical size of the object is varied, and the response frequencies calculated. When the calculated and observed resonant frequencies match, the object's size is that value which was used in the calculation of the matching theoretical solution.

[0040] By way of the methods of this invention, once the presence of such a response (i.e., late arriving resonant waves) is detected, the next step is to determine the object's exact position. This is done at step 112 by analysis of the seismic response patterns. Assuming that the seismic propagation velocity of the ground immediately around the buried object is fairly constant, it follows that the sensor directly above the buried object will be first to detect the resonant waves emitted by the object, the time of receipt of the same emitted wave by an adjacent sensor taking longer due to the greater distance of the object from the sensor. When represented visually, a time of arrival plot manifests as a hyperbolic curve, with the trough or inflection point of the curve coinciding with the sensor positioned closest to the buried object. For the middle sensor positioned directly over the buried object, the plot will manifest as a symmetrical hyperbolic curve with its inflection point at the center of the receiver's line as illustrated for the Example, Figure 5, shot position 51. [0041] As the time of travel of the resonant waves to the sensor overhead is affected by the distance from the resonant object to that sensor, once the lateral location of the buried object is established, its depth can be readily calculated. By way of example, and with reference again to Figure 5, taking into account the evaluated propagation velocity V_s for the shear waves of the embedding materials, known as a result of measurements for propagation of direct waves from the source, the closer to the surface the buried feature is positioned, the sharper will be the inflection of the hyperbolic response. With the aid of a software program to calculate the responses that would be expected at various depths, a vertical model of responses as a function of depth is computed, Step 112 of Figure 1, and the data obtained in the field then matched to the model in order to approximate the depth of the detected feature, Stepl H of Figure 1.

[0042] In terms of the solution for localization (imaging) of detected resonant objects, a knowledge about distribution of background shear waves velocity F₄ is required, which allows computation of shear wave propagation times τ_ιk between any underground point M_k = (x_k,y_k ,z_k) and sensors locations r_t . Information about velocity F₄ can be obtained from analysis of the direct or reflected waves, or from some other seismic methods. This information can also be obtained by performing inversion of resonant waves using a range of inversion velocities F₄ where correct value of F₄. provides a localized sharp image.

Modeled emissions W_lk (/) generated by a source at point M_k , and recorded in the sensor location r_t , are evaluated using the formula

W_lk (J,_a ) = &φ(i2πf_ιaτ,_k )/ p_lk ,

where p_lk is a wave travel distance between points M_k and η . Points M_k which provide a best fit of the modeled field data W_lk (/_re4 ) with transformed recorded data S₁ (f_ιes) localize the object. For example, localization can be done using maximums of imaging function

F(M_k) = ∑ ∑s;(f,_esW_lk(f_res) where s ^*(f_res) is a spectral component of data from shot/, recorded in receiver i

(l ≤ i ≤ I). The asterisk (*) denotes a complex conjugation. Summation covers all shots in the vicinity of the imaging area. A software program is used to perform the required calculations of the above formula, the calculated value of F(M_/J displayed as a function of spatial coordinates. At the true object location, the summation goes constructively, sharply increasing the value of the imaging function. At all other points, summation is destructive resulting in lower image values. With reference to Figure 6, this manifests in the image depicted, the maximum imaged function corresponding to the true location of the object. Such imaging does not require visual analysis of recorded traces. Use of absolute values in equation for F(M_k) makes the inversion independent of shot origin times and stabilizes the inversion result.

[0043] In an alternative embodiment of the invention, a passive system is provided where ambient seismic waves are relied upon to provide the source of seismic excitation. In this embodiment, for an area of interest, a line or grid of sensors can be laid out, either above or below the ground surface. These sensors in turn can be linked to a seismic monitor by wire or wirelessly. Any number of ambient sources can generate a seismic wave that can cause a buried feature of impedance contrast to resonate and the resonant wave detected. Such seismic source could include a low intensity earthquake, a movement of vehicles in the vicinity of sensor grid, or active excavation underground in the area of the sensors, running machinery and vehicles, wind, walking people.

[0044] Yet another embodiment of this invention is illustrated in Figure 2. Shown are two underground features, with resonant waves emanating from each in response to an incident seismic wave. The array of source and receivers (that is, sensors) shown at ground level is illustrative of the first embodiment described above. In this variation, at a fixed location, a borehole can be drilled, and a controlled source and a line of receivers placed into the hole. Shot gathers are taken by activating the source. This modality can be used in combination with surface examination, in that once the lateral location of a feature is determined, the borehole can be drilled, the controlled source and sensors placed. With these additional readings, the depth of an object can be accurately fixed using the same methodologies as described above.

Example

[0045] A field example will now be described. The experiment was conducted to generate a test data set using the methods of this invention for subsurface object detection and imaging. The soil comprised compacted, consolidated sand with a measured P- velocity of 500 m/s and S- velocity of 240 m/s. The test object was a water-filled barrel 60 cm in diameter and 120 cm long. Some air bubbles were left inside the barrel before it was sealed. The barrel was placed at a depth of 5-6 meter in a specially excavated pit which was then back filled with soil. An in-line channel of 24 vertical geophones, spaced 0.5 m one from the other was used for every ground shot made by a sledge-hummer. The shot point was positioned at the first receiver of each line. There were 87 shots in total with 1 m spacing intervals (i.e. the line of sensors was moved in 1 m increments). The middle receiver of line #51 (i.e. shot gather #51) was directly above of the barrel. Records were 0.5 seconds in length with a 0.5 ms data sampling rate. Figure 3 shows raw shot gathers #48-57 Direct P- and S- waves had dominating amplitudes on each trace, and provided no useful data for determining the location of the object. After applying automatic gain control (AGC) the late (0.25-0.5 ms) interval revealed sharp resonant peaks in amplitude spectra at 78 Hz (Figure 4). Band-pass filtering of traces in the 75-83 Hz frequency interval gave a clear signature of a secondary seismic source with an apex in the middle of the line 51 (Figure 5). The phase slopes (240 m/s) of the secondary source correspond to velocities V₃ of shear waves in the embedding medium. The result of migration of data from Figure 5 is shown on Figure 6 providing an accurate location for the barrel.

[0046] Fitting a first resonant peak for a liquid-filled sphere with radius R=O.3 cm gave a fluid velocity 70 m/s which is consistent for a fluid with air bubbles model with 1 g/cm³ density (Figure 7).

[0047] The field data is modeled with 2D finite-difference seismic wave propagation program using a 4Ox 12.5 m model with 0.05 m grid spacing. Background Vp and Vs velocities have the same values as in the field experiment. The local object was modeled as a 1 m x 0.7 m rectangular containing liquid with 70 m/s velocity and 1 g/cm³ density. Ten shot gathers were computed using 3 m spacing for sources. The surface receiver line was the same for all shots and had 61 sensors separated by 0.5 m. A single shot gather for the shot #3 is shown on Figure 8, where similar to the field case, primary scattered waves rapidly decay after 0.2 seconds. The recording length was 1 second.

[0048] A snapshot image at 0.3 seconds (Figure 9) shows that at a later time the propagating field is predominantly comprised of shear waves caused by circumferential waves in the object. Spectral content of the late parts of the records (after 0.5 s) shows the presence of a sharp peak at 53 Hz in all traces (Figure 10). AU the data were automatic gain controlled and band-pass filtered around the peak frequency leading to the images like that shown in the Figure 11 and revealing a quasi hyperbolic phase arrivals similar to the one based on real data (Figure 5).

[0049] The methods of this invention capitalize on the observation that high-contrast subsurface heterogeneities trap seismic energy and are capable of releasing that energy long after the recordings of primary scattered waves. This trapped energy primarily consists of circumferential waves propagating along the perimeter of the object and radiate into the surrounding medium as slowly decaying body waves. The circumferential waves propagate in all directions around the object which potentially can create discontinuous travel time curves. Most of the energy irradiated back to the embedding medium is carried by shear waves. This simple data pattern makes it attractive candidate for inversion techniques based on back-propagation principles where recorded traces are used as sources in reverse time, and a suitable program used to "collapse" the waves back to a point of resonant source origin. It is expected that multiple objects can be detected and imaged using separate resonant frequencies. It is clear that exact timing of the source excitation is not important and trapped energy radiation can possibly be observed in the presence of the background noise, leading to cost effective object detection techniques.

[0050] This invention has now been described herein in considerable detail to provide those skilled in the art with information relevant to apply the novel principles and to construct and use such specialized components as are required. However, it is to be understood that the invention can be carried out by different equipment, materials and devices, and that various modifications, both as to the equipment and operating procedures, can be accomplished without departing from the scope of the invention itself.

Claims

I CLAIM:

1. A method of determining the location an underground feature, comprising, providing a wave to a ground; and detecting for a resonance emission wave in said ground; such that the location of said underground feature is determined.

2. The method of claim 1, wherein said providing step comprises generating a wave from a source on the surface of or under said ground.

3. The method of claim 1 , wherein said underground feature is a solid body, or a fluid containing body.

4. The method of claim 3, wherein said solid body is a buried object.

5. The method of claim 3, wherein said underground feature" is a fluid filled cave, UXO, tunnel or a fluid filled fracture.

6. The method of claim 1, wherein said detecting step comprises providing a plurality of receivers capable of detecting said resonance emission wave.

7. The method of claim 6, wherein said receivers comprises one or more receivers on the surface of the ground.

8. The method of claim 6, wherein said receivers comprises one or more receivers under the ground.

9. The method of claim 6, wherein said resonance emission wave is a shear wave.

10. A method for determining the underground location of a high contrast resonant object embedded in an elastic medium comprising: providing a plurality of sensors for monitoring seismic waves; providing a seismic source; generating a primary wave using said seismic source; recording the seismic response for secondary, resonant waves generated by said resonant object; and, thereafter determining which of said sensors is closet to said object, whereby the approximate location of said object is established.

11. The method of claim 10, wherein the seismic source and sensors are arrayed on the surface of the ground.

12. The method of claim 10, wherein a seismic source and sensors are arrayed underground.

13. The method of claim 10, wherein the location of the underground object is determined by calculating anticipated response times for emitted resonant waves at various positions underground and fitting these calculated response times to the response times measured by said sensors.

14. The method of claim 10 wherein the seismic source comprises ambient noise.

15. The method of placing a resonant object underground to excite it from a distance at resonant frequencies using active controlled sources.

16. A method for determining the location of a below ground buried high contrast feature, the method including the steps of: a) providing a seismic source; b) deploying a plurality of seismic sensors in a predetermined pattern; c) generating a seismic wave; d) detecting the seismic responses to said seismic waves of said deployed seismic sensors, and recording said detected data; e) analyzing the pattern of responses for the presence of emitted resonant waves; f) identifying the approximate location of the high contrast feature based on said analyzed response patterns; g) calculating theoretical response data as a function of an assumed location of the buried feature, including both its lateral position and depth; and, h) thereafter matching the calculated theoretical response data with recorded detected data to determine the location of the underground feature in three dimensions.

17. The method of claim 16 wherein the emitted resonance waves are analyzed for the presence of resonant peaks at the same response frequency.

18. The method of claim 16 wherein the steps of providing a seismic source and generating a seismic wave comprises the single step of using ambient seismic waves.