A das-based seismic imaging method

Abstract

Described herein is a method for performing a seismic survey, comprising: positioning at least two fiber optic cables and defining a target area that is located in between the cables and delimits the top of a sub-surface target volume; coupling an end of each cable to a distributed acoustic sensing interrogator; and activating one or more seismic sources exclusively outside of the target area. Also described herein is an apparatus for carrying out a seismic survey.

Description

[0001]A DAS-Based Seismic Imaging Method The present invention relates to a method for seismic imaging using DAS technology, and in particular to a method using multiple DAS cables and a sparse source setup to image surface waves in a volume between the cables. Surveying techniques are an important tool in determining the structure of the subsurface of the earth. The information obtained from such surveys is of wide scientific interest and can be used to study the make-up and history of the earth and the subsea environment, as well as for locating and accessing underground reservoirs of hydrocarbon-based fossil fuels. One type of survey using sensors for studying the structure of the earth’s subsurface is a seismic survey. A typical seismic survey uses one or more seismic energy sources which are most often active sources, but can be passive, in order to generate an elastic wave signal which propagates into the earth’s subsurface. The properties of the elastic wave signal and / or its direction of travel is altered depending on the properties of the subsurface region through which it passes. The signals are then detected and recorded by seismic receivers located at or near the surface of the earth, and can be processed to recover information relating to the lithologic subsurface formations, identifying such features as, for example, lithologic subsurface formation boundaries. Those skilled in interpreting such data can infer the presence of oil or gas within the subsurface or can study how an injected CO2 plume develops and moves / migrates within the subsurface. The seismic data can also be used to derive information about subsurface properties which can be valuable for those carrying out construction work. The data can inform as to where and how to place foundations for wind turbines, anchoring sites, and pipelines, or can assist in searches for seabed minerals and in inferring the presence of shallow gas, the latter being relevant in terms of risk reduction (QHSE). In a marine environment the seismic receivers are often towed behind a vessel in long cables known as streamers, or they are placed on the seafloor as nodes or seabed cables. The use of nodes or cables located on the seafloor is both costly and time-consuming, but is still often preferred since stationary sensors on the seafloor can record high fidelity reflection data and the data collected is subject to lower levels of noise. Receivers have traditionally been made up of hydrophones, but in recent years other sensor types such as geophones, accelerometers, and various forms of MEMS sensors have increasingly been deployed in order to supplement or replace the hydrophone measurements. Fiber-optic cables can also be used as seismic sensors. Distributed Acoustic Sensing (DAS) technology utilizing fiber-optic cables encompasses several techniques designed to replace individual, discrete measuring devices with what is essentially one long continuous sensor. Each of these techniques involves hooking up the end of a fiber-optic cable to a device known as an interrogator, which sends out a continuous train of laser pulses. The very faint reflections generated as each pulse is scattered at multiple points along the fiber-optic cable, corresponding to impurities in the cable, are then recorded. This interrogator in such a case is often referred to as a DAS interrogator. The method relies on the phase changes resulting from Rayleigh scattering of the laser pulses from naturally occurring structural defects present throughout a standard optical fiber. The idea is that any seismic wave will slightly stretch and compress the fiber as it passes through the ground within or on which the fiber is located. The phase of the light pulse at each scattering point corresponds to the variation in fiber length at that point on the cable. As such, each point acts like a tiny virtual microphone (hydrophone) or seismometer, whose distance along the fiber is established by measuring the time it takes the reflected light to complete its round trip. With pulse frequencies in the kilohertz range and the virtual sensors being spaced as little as a meter apart, the technique provides a potentially very fine- grained record of strain variation along the fiber with respect to both space and time. Through signal processing, the data collected can be used to produce a seismic trace with subsurface reflection energy similar to that produced by more traditional sensors. DAS and purpose-built fiber-optic systems have in recent years been successfully deployed on several oil fields in connection with what is referred to as Permanent Reservoir Monitoring (PRM) systems and borehole seismic systems. Where DAS cables have been used to study the structure of the subsurface, this has involved the use of a single cable in order to study reflected and refracted waves (see, for example, the studies by Raknes et al.2023 and Bachrach et al.2023), or a single cable in a snaking configuration. In the first case only a 2D map of the sub-surface is achievable. Suitable source configurations have not been considered in detail. Improvements to surveys using DAS cables as the seismic sensors are desirable. According to a first aspect of the present invention, there is provided a method for performing a seismic survey, comprising: positioning at least two fiber optic cables and defining a target area that is located in between the cables and delimits the top of a sub-surface target volume; coupling an end of each cable to a distributed acoustic sensing interrogator; and activating one or more seismic sources exclusively outside of the target area. This survey setup comprises multiple DAS cables, each having a DAS interrogator connected at one end, and positioned to at least partially surround an area on the surface delimiting a sub-surface volume to be imaged. The target area is contained in the area between the cables, in that it does not extend outward of at least the outermost cables in a horizontal direction (and in some examples may not cross or extend outward of any of the cables). Where survey design and the type of interrogator system used allows, multiple DAS cables may be interrogated by a single interrogator unit so that the ends of two or more of the cables are coupled to the same unit. Alternatively, some or each of the cables present can be coupled to their own interrogator unit. Reference to seismic sources being activated outside of the target area is to them being activated at positions which are not directly above the target area, but which are offset from this area in a horizontal direction. This relative arrangement of the cables, the target area, and the source activation positions is extremely efficient in terms of both time and equipment required to carry out the survey. The relative positioning allows surface waves travelling between two of the DAS cables to be imaged in the best way, with minimal influence on collected data from other types of waves (reflected or refracted waves, for example), particularly after processing using interferometry. This use of surface waves for imaging is advantageous because surface wave signals will not typically be aliased along the fiber direction due to the quasi-continuous spatial sampling of seismic traces along the fiber. The method is able to image the sub-surface down to at least a few 100s of meters deep, such as down to around 500m depth, depending on parameters of the survey. The usual source carpet is replaced with one or a small number of source lines which are still able to produce useful data when their placement relative to the DAS cables is as specified above. The use of multiple DAS cables also results in reduced cost and improved signal quality as compared to a single cable survey. Cables are easier to lay, and far fewer source activation positions are required (a sparse source layout rather than a source carpet is possible), both of which can be crucial in the context of seismic surveys where operational costs are high. Accurate modelling / characterization of the near surface which focusses on shear- wave near-surface structure, and which is relatively computationally efficient and inexpensive to achieve as set out above, will be valuable for many applications relating to carbon capture and storage, offshore windfarm geotechnical assessment, deep sea mineral exploration, and hydrogen storage monitoring, among others. The method could be used to find suitable sites for CCS or hydrogen storage, or for monitoring progress at sites which are in use. Knowledge of sub-surface structure will also be invaluable in selecting the best sites for the construction of offshore windfarms. In some cases, these methods could act as a screening prior to more traditional methods to be used in cases where anomalies are detected. In embodiments, the same interrogator can be coupled to the ends of two or more of the cables so that the cables “share” an interrogator. Alternatively, each cable can be equipped with its own, separate, interrogator. Each cable will, however, have an interrogator coupled to it at one end, so that signals from that cable are received at the interrogator. A DAS cable (distributed acoustic sensing cable) refers to an optical fiber cable which is configured to function as a sensor within a system including an interrogator. The interrogator is an optoelectronic device which sends optical signals into the optical fiber cable from one end, receives optical signals from the cable through the same end, and converts the received optical signals into electrical signals. Reference to defining a target area between the cables is to the target area being located entirely within the area surrounded or partially surrounded by the two or more cables, or the area located in between the two or more cables. The area and the cables may be located on the surface of the earth, such as on the seabed, or may be located at the same depth a short distance under the surface. In general, the cables and the target area will be within the same horizontal plane and references to “within” and “outside” are to relative placements in a horizontal direction. Although source activation positions can also be offset vertically from the target area (they can be above a target area of the seabed towed by a vessel, for example) reference to positions being outside the target area is to the fact that the activation positions are outward of the target area in a horizontal direction. The activation positions will also generally be exclusively outward of the DAS cables themselves. The target area can be smaller than the area between the cables, or can extend right up to each of the cables in some cases to completely fill the area between the cables. The target area is defined as the area delimiting a target sub-surface volume from which signals are both received and processed to provide a representation of that volume, which may be a 3D representation. The target area thus defines and delimits an upper extent of a sub-surface volume. Either way, the survey is designed so that the target area is contained completely within the area of the surface that is located in between the cables, and does not extend outward of this. Any subsurface signals from outside of the target area (or the target volume defined by the target area) are simply ignored, or are treated as noise, during subsequent data processing, and are in this way suppressed from the data used for imaging. Source lines also at least partially surround the target area, but are located further from the target area than the cables in a horizontal direction. The source lines can also obviously be offset vertically, and will usually be so in the case of a seabed seismic survey. Sources may be activated at spaced intervals along the entire length of a source line. The intervals between shots in a direction along the source line can be evenly spaced in terms of distance or irregularly spaced. The source itself can be any kind of active seismic source, but will preferably be a marine vibrator, sparker- type source, and / or air gun for a marine seismic survey. If a marine vibrator is used, the position of the shot on the source line will represent the start of each sweep. In embodiments, activating the one or more seismic sources comprises activating the seismic sources along at least one source line, wherein the at least one source line is offset horizontally from the perimeter of the target area along the entire length of the source line, such that the at least one source line follows the shape of the target area perimeter and at least partially surrounds the target area. Source lines will this way surround and define a larger area, also containing the target area. The source lines are horizontally offset from the region by some distance, and are located outside of the target area. This setup allows for monitoring of surface waves travelling from all (or at least more) points on one cable to all points on another cable, whilst still minimizing time taken to complete the survey. In embodiments, the at least one source line extends along or around the target area so that at most two source lines are encountered when travelling in a direction outwards from the target area (i.e. in a straight line outwards from the target area). This refers to positions of the target area and sources in a single horizontal plane, as viewed from above. The number of source lines crossing one another when viewed from above is generally kept to a minimum, and the survey can be efficiently performed. In embodiments, the at least one source line extends along or around the target area so that only a single source line is crossed when travelling in a direction outwards from the target area. Again, this refers to relative positions when viewed from above. The at least one source line may consist of a single source line. The single source line may completely surround the target area. The source lines used for the survey will not be repeated at intervals to form the usual carpet or grid of source points. The source line or lines extend instead around the target area as one single closed or open loop, so that any line extending outward from the target area will only cross a single source line. When combined with at least two DAS cables located on either side of a target area and used to image surface waves, a single source line outward of the target area may produce an accurate 3D representation of the target volume without the need for additional source lines. In embodiments, the at least one source line extends along or around at least two opposing sides of the target area. Surface waves travelling in both directions between the two cables can in this way be detected and thus the surface wavefield is better sampled, which results in a better-quality representation of the sub-surface once the collected data are processed. In embodiments, the method comprises placing two fiber optic cables either side of the target area and activating the one or more sources along two source lines extending along opposing sides of the target area substantially parallel with the two fiber optic cables, and outward of the cables. The two source lines will therefore be located further from the target area in a horizontal direction than the cables. In embodiments, the at least one source line is offset horizontally from the target area perimeter by between 500 meters and 2000 meters along the entire length of the source line. Maintaining the source line at a distance from the target area ensures that it is predominantly near-horizontally-propagating surface waves which affect strain in the cable, rather than near-vertically-propagating reflected or refracted waves. In embodiments, the at least one source line extends along at least two thirds of the length of the target area perimeter. In embodiments, the at least one source line completely surrounds the target area. In embodiments, activating the seismic sources comprises activating the sources at locations which define a single open loop or a single closed loop. In embodiments, activating the seismic sources comprises activating the sources at locations which define an open loop extending through at least 270 degrees around the target area. In embodiments, the method comprises acquiring, by the interrogator coupled to that cable, data from each of the cables. In embodiments, acquiring the data comprises acquiring data relating to strain in the cable caused by surface waves travelling between the cable and another cable across the target volume. In embodiments, the method comprises acquiring, by the interrogator coupled to that cable, data from each of the cables simultaneously with activating the one or more seismic sources. In embodiments, the method comprises processing the acquired data to produce a 3D representation of the sub-surface target volume. The processing of the data focusses on the effects of the surface waves on the cables, rather than other types of acoustic or elastic signals. In embodiments, processing the data comprises minimizing a part of the data signal resulting from reflected and / or refracted waves. In embodiments, processing the data comprises minimizing or removing a part of the data signal that results from other than surface waves. The influence of other types of waves on the processed data signal is minimal or is negated, so that only or substantially only the contribution from surface waves remains. In embodiments, the processing comprises processing the acquired data using interferometry. This allows surface waves to be effectively emphasized in the resulting data, with influence on the signal from other types of waves being excluded. In embodiments, the at least one source line extends along at least two thirds of the length of the target area perimeter. In embodiments, the at least one source line completely surrounds the target area. This is advantageous in that the sampling of various signals passing through a cable from different regions of the other cable or cables is maximised. In embodiments, activating the one or more seismic sources comprises activating the sources along a source line which defines a source area, wherein the source area completely contains the target area. In embodiments, acquiring the data comprises acquiring passive data during a period in which the one or more seismic sources are not activated and acquiring active data simultaneously with activating the one or more seismic sources. The sources can comprise marine vibrators in some embodiments, or another type of active seismic source such as an air gun which is designed to produce a controlled signal at a particular frequency for seismic surveying, and these may be towed or directed along a source line from one end to another whilst being activated at intervals. In embodiments, the method comprises, after acquiring the passive data and before acquiring the active data, determining signal bandwidth and / or signal-to-noise (SNR) levels from the passive data and selecting one or more of the shot positions and spectral properties (e.g. amplitude and / or phase characteristics) for the one or more seismic sources to be used during active data acquisition based on the determined signal bandwidth and / or signal-to-noise levels. The active data acquisition can in this way be tailored to better complement the passive data. In embodiments, the method comprises: combining the active and passive data; producing a pseudo-source dataset from the combined data using interferometry methods; and producing a 3D representation of the sub-surface target volume from the pseudo-source data using surface wave inversion. The active and passive data are therefore processed together as a combined data set, rather than producing separate maps from each. In embodiments, the method comprises selecting the frequencies emitted by the one or more seismic sources so that the active data contains a partial spectral overlap with the passive data and the frequencies emitted by the one or more seismic sources extend beyond those present in the passive data. The source survey design parameters are therefore chosen so that the active source data contains a partial spectral overlap with that of the passive data, and the active source bandwidth extends beyond that of the passive data. The bandwidth overlap will enable processing to achieve an optimal blend of active and passive signals in a single dataset, while the added bandwidth will ensure that the combined data carries more information about the subsurface than either the passive or active data alone. In an example, the passive data spans a range between 0.1Hz and 20Hz, and the active data spans from 5Hz to 1kHz. In embodiments, therefore, the overlap range lies in the range between around 1 Hz and around 50Hz, more preferably in the range between 5Hz and 30Hz, and most preferably in the range between around 5Hz and around 20Hz.The overlap may span the entire frequency range quoted in any of the examples provided above, or may span only a part of the range. In embodiments, the method comprises activating the one or more seismic sources along at least one source line which extends along the perimeter of the target area on at least two opposing sides of the target area. Surface waves travelling in both directions between cables on either side of the target area can in this way be detected and sampled. In embodiments, the at least two fiber optic cables define the perimeter of the target area. The target area therefore extends right up to the cables on either side. In embodiments, coupling an end of each cable to a distributed acoustic sensing interrogator comprises coupling an end of each cable to a different distributed acoustic sensing interrogator. In embodiments, the survey is a marine seismic survey. According to a second aspect of the present invention, there is provided an apparatus for carrying out a seismic survey, comprising: at least two DAS cables positioned outside of a target area that is located in between the cables and delimits the top of a sub-surface target volume; an interrogator coupled to an end of each cable; and at least one source line defined as a pathway along which one or more seismic sources for the survey are to be activated, wherein the at least one source line is exclusively outside of the target area. In embodiments, the at least one source line is offset horizontally from the perimeter of the target area along the entire length of the source line, such that the at least one source line follows the shape of the target area perimeter and at least partially surrounds the target area. According to a third aspect of the present invention, there is provided a method for performing a survey comprising: positioning at least two DAS cables to define a target area that is located in between the cables and delimits the top of a sub- surface target volume; coupling an interrogator to an end of each cable for the collection of data from the cables; collecting passive data from the at least two DAS cables during a passive period when the seismic sources are not being activated; activating seismic sources along at least one source line during an active period and collecting active data from the at least two DAS cables during the active period; combining active and passive datasets and producing a pseudo-source dataset from the combined data using interferometry methods; producing a 3D representation of the sub-surface target volume from the pseudo-source data using surface wave inversion. In embodiments, the method comprises, after collecting the passive data and before collecting the active data, determining signal bandwidth and / or signal-to-noise (SNR) levels from the passive data and selecting the one or more of the shot positions and amplitude for the active sources to be used during the active data collection based on the determined signal bandwidth and / or signal-to-noise levels. Embodiments of the present invention will now be described, by way of example only, with reference to the following figures wherein: Figure 1 is a plan view of a survey setup showing a sub-surface target volume, a target region, and cable and source positions, Figure 2 shows the survey configuration of figure 1 as a side view, Figure 3 shows the survey configuration of figure 1 from above, including the source line and cable positions, and Figure 4 shows alternative positioning for the cables and source lines. The setup and methods described herein allow for the use of DAS cables and interrogators to map a region of the sub-surface accurately by detecting predominantly surface waves travelling between two distinct cables located either side of a sub-surface volume of interest. Surface waves travel more slowly and penetrate less deep into the sub-surface than reflected and refracted waves. There are two main types of surface waves (Love and Rayleigh), which can be distinguished based on the direction of motion of particles as the wave passes through a region. Both of these types travel along the earth’s surface or in the relatively shallow sub-surface, and reduce in amplitude rapidly as the distance from the surface increases. One possible configuration for a survey which is optimized for the detection of surface waves is shown in figure 1. In this case, two DAS cables 6 are placed side by side on or near to the earth’s surface, such as on or near to the seabed. The cables 6 can be arranged to sit on the earth’s surface, or can be buried a small distance under the surface, for example in shallow trenches, in order to try to improve coupling. In the example shown in figure 1, the cables run generally (sub)parallel to one another and delimit an area of the earth’s surface in between the two cables 6 as shown in the figure, and extending as far as the ends of the cables at either end to form a substantially rectangular target area 2. The volume of the sub- surface located just below this area 2, and extending from the delimited area downwards a given distance, is the sub-surface target volume 4. Data relating to seismic waves passing through this volume will be collected during the survey using both DAS cables, and will be processed to produce a 3D image or map of the target volume including information about its properties and structure. Surface waves passing between the two cables in either direction will be detected at both cables, allowing for the interference / estimation of 3D images or models. Although the cables do not necessarily need to extend in straight lines, processing in this case may be modified to simply account for the correct positions of all points along the multiple DAS cables, under the assumption that the geometry and coordinates of each DAS cable are captured during laying of the cables and data acquisition. Figure 1 also illustrates possible source lines 8 for active source collection using the two cables laid out in a parallel configuration. The shot / source lines are positioned some distance away from the cables 6 so that they are offset in a horizontal direction from the cables (far-offset sources), but mirror the direction in which they extend so that they still run parallel to the cables and parallel to the perimeter of the target area on two sides. It would be possible to lengthen one or both of the cables to extend also around the top and bottom edges of the target area 2 (the nearest and furthest edges of the target area is shown in figure 1), or to use one or more additional cables for this purpose. In this case, therefore, the source lines 8 can also be extended to turn through 90 degrees in order to run parallel to the target area 2 along the top and bottom sides of the target area perimeter. The source line(s) in this case may form a shape that reflects, or is the same as, the shape of the target area (in this case a rectangle). The shape traced by the source line(s) will be larger than, and will surround, the target area. In all cases, the source lines will be offset from the perimeter of the target area 2 by at least a predetermined horizontal distance in order to ensure that it is surface waves, rather than reflected or refracted waves, that are primarily detected at the cables. Surface waves travelling between the two cables 6 are detected due to the strain experienced by the DAS cables as a result of the waves passing. Effects from passing reflected and refracted waves are also present but it is essentially only surface waves which remain after processing, due to the acquisition geometry and survey design described above. The processed data is used to produce a representation of the subsurface down to a few 100s of meters below the surface (within the target volume 4) based on surface wave effects. The interrogator is configured to measure a change in strain, which is converted to a digital representation for further transmission or for storage on a data storage device. Processing of these signals to produce a representation of the sub-surface target volume can be carried out by a processor that is collocated with the interrogator, or can be passed to an external processor to produce the images or representations of the sub-surface at a remote location. Interrogators are located at the ends of each of the DAS cables used for the survey. As mentioned above, and as shown in figure 4A, the interrogator at the end of two or more of the cables can be replaced with a single interrogator 12 serving two or more of the DAS cables 6 present. Each of the separate cable sections will in all cases have an interrogator coupled to one end, and will as such represent a distinct DAS cable. Figure 2 shows the same survey setup as illustrated in figure 1, but as a side view. In this case the survey is a marine seismic survey and the DAS cables 6 are located on the seabed, with the sub-surface target volume 4 being located directly beneath a target area 2 between the cables 6. Near offset sources are shown for comparison with conventional seismic surveys – whether using DAS or other sensor types - only. The horizontal offset between the source lines 8 and the cables 6 will be within a certain range for all points along the length of the source line, but parameters of the survey can be selected to ensure that, after processing, the majority of signals received from within the target volume 4 are due to surface waves, rather than waves that have been reflected or refracted within the target volume. Processing methods focus on waves that have travelled from one of the cables 6 to the other, i.e. that have travelled from the source, through one of the DAS cables, through the target volume 4, and have been received at another of the cables on the opposite side of the target volume from the source. For active data with source acquisition geometries designed as proposed, the signals retrieved after processing will include predominantly surface waves travelling through cables on either side of the target volume. It is preferable that the source lines 8 extend along two opposing sides of the target volume, at a horizontal offset as described above, so that surface waves travelling in both directions between the cables can be properly sampled during the survey. The horizontal offset for the sources can be selected in order to prioritize the retrieval of these surface waves, while suppressing other wave types, after processing. An optimal offset will depend on the size of the target area and the height of the source lines above the earth’s surface (the vertical offset between the source line or lines and the target area). The contribution of reflected waves to the output of interferometric processing by near-offset sources is illustrated in figure 2. Here reflected waves from within the target volume pass through a cable on one side of the target area, are reflected within the target volume, and are detected by a cable on the other side of the target area. Given the depth of the target area in this case, placing the source lines only at a larger horizontal offset (as for source lines 8 shown in the figure) will mean that, after interferometric processing, any reflected waves will not be retrieved because near-offset sources are absent. Only surface wave signals sampling the target volume between cables will be accurately retrieved, these necessarily have been reflected from a position closer to the earth’s surface. The survey design described herein allows enough information to be retrieved to infer 3D elastic subsurface models with minimum source acquisition / operational effort. In general, the cables are set up to detect and enable the interferometric retrieval of surface waves which have travelled through the sub-surface between a DAS cable on one side of the target area to a DAS cable on the opposite side of the target area. This is illustrated in figure 3, which shows how the signal from active sources 10 is detected at both cables 6 before and after passing through the sub-surface target volume. Source lines 8 are located on either side of the target volume, so cables will pick up surface waves travelling in both directions. Properties of the source or sources can also be adjusted in order to optimize the response of the survey to surface waves in terms of both desired bandwidth and SNR. The frequency content of the source will affect how deep the surface waves detected at the DAS cables penetrate. For lower frequency sources, the surface waves will penetrate deeper, which can increase the depth of the sub-surface target volume. In general, passive data tends to be at a lower frequency and can therefore penetrate deeper into the sub-surface than data collected during, and as a result of, the use of active sources. Complementing active data with passive data can greatly improve the results of the sub-surface imaging, and these can be combined in a novel way for this type of survey, as is explained in more detail below. Where both passive and active data are collected, the passive data are ideally collected first (although this is not necessarily the case, and will depend on operational constraints). The collection of passive data involves the detection of wave signals – which contain surface waves - using the DAS cables in cases where the surface waves originate from natural seismicity, ocean and weather-borne noises, and / or other background sounds which are not a result of the use of nearby active sources. The time period during which passive data are collected will most often be of similar length to the subsequent or preceding time period required for the collection of active data, but may be shorter or longer in some cases due to the varying level of noise in the passive data. The passive sources will generally produce wavefields having a lower frequency than is achieved via nearby active source activation. In terms of temporal bandwidth, as a non-limiting example, passive noise generated by weather and oceanographic phenomena , combined with nearby human activity (e.g., oil platforms or other subsea installations) typically emits surface wave signals between 0.1-20Hz, while combining different types of active seismic sources can reliably span from 5Hz to up to 1kHz or higher. Once the passive data have been collected, or in some cases before, active data are collected using the same cable layout but activating sources along the source line or lines which surround the target area at a horizontal offset. The active source properties can be designed to complement the passive data. The active data may be of a higher frequency, for example, in order to extend the bandwidth of the combined data as far as possible to improve the achievable resolution. For example, if the objective is to be able to resolve subsurface model features within a 10m horizontal resolution or greater over a 5km by 5km area, assuming a near-seabottom surface- wave wavespeed of 500m / s, it would be necessary to sample at least a bandwidth approximately between 1.5-100Hz to achieve a ±1m uncertainty. This can be achieved by combining the active and passive data. Once both the passive and the active data have been collected, the data are processed to produce an image or model which contains information relating to properties of the sub-surface target volume (a representation of the target volume). In most cases, suitable pre-processing is first applied to both the passive and the active data, i.e. filtering, windowing, and so on. Following pre-processing, seismic interferometry methods are used to determine properties of “pseudo-sources” and a set of pseudo source data are produced for both the passive and active data collected using the DAS cable setup. Pseudo source datasets represent surface wave responses between cables which are reconstructed to reflect a situation in which pseudo-active sources are present along each cable, at discrete intervals, and the wavefield from each pseudo source is detected along all other cables present. The pseudo source datasets have the same sampling as the original DAS recordings, which are discretized but very finely sampled so that they can be considered more-or-less continuous. In a preferred example, the passive and active datasets are first combined and then jointly processed using a suitable interferometry method to yield a single pseudo- source dataset which is based on both the active and passive data. A 3D model or image of the sub-surface target volume can then be produced from the pseudo- source dataset by applying a suitable surface wave inversion method to this dataset, such as full wave inversion, tomographic inversion methods, or an eikonal (or any other traveltime-based) method. As an alternative, a subsurface model can be produced from one of the passive or active datasets first, by creating a pseudo- source dataset using interferometry and then a model using inversion. This can then be refined using the pseudo-source data from the other data set, also produced using interferometry, or using a model produced from the other pseudo-source dataset using inversion. Joint processing is preferable and will allow inversion methods to resolve for full 3D models (all parameters at once) while also taking into account all pseudo shots at once, derived from both passive and active data. The below sets out one possible example of the processing of both passive and active data into single pseudo-source data. Other methods can be used. For any two receiver locations and ^^2(i.e., point-wise DAS-measured time-series) , each on separate DAS cables, active and passive data are acquired. Active data ^^^^for each receiver location are taken for corresponding active shot locations ^^^^, thus (^^^^)and ^^^^^^2(^^^^). Passive data are first analyzed for the characteristics of noise types and source / excitation categories. This is achieved by conventional signal processing approaches such as time-frequency analysis, filtering / beamforming, and so on, which are applied to the DAS passive data prior to receiver selection for pseudo source processing. Following noise analysis, the passive data ^^^^are re- arranged into time-gathers ∆^^^^, by a combination of time-windowing, filtering, and grouping / selection, according to their common noise characteristics, with information about the noise (predominant source directions, frequency bands, etc.) kept as meta-data to inform subsequent steps. Assuming a data representation in the frequency-domain, the first step towards pseudo-source is to compute the following quantities: where here the quantities (^^^^^^1 (^^^^), (∆^^^^)) denote active and passive data for a given receiver (^^1) at fixed frequencies ^^^^, while (∆^^^^)) are their counterparts for another given receiver. At each frequency, the matrices (^^^^^^^^, ^^^^^^^^) are 4x4 weighting matrices that aim at balancing the contribution of active and passive data toward the desired pseudo-source data at given frequencies. Because the * superscript denotes complex conjugation, the operations correspond either to (weighted) auto- or cross-correlations of the various passive or active data components. In a next step, the quantities above are combined into the quantity: where the brackets represent averaging / stacking over all chosen (^^1and ^^2) pairs. This calculation represents a modified version of seismic interferometry by deconvolution, and thus the quantity (^^^^^^) is the desired single pseudo-source response (at a fixed frequency) between the receivers ^^1and ^^2, provided that the weighting matrices are set appropriately. In practice, however, while the spectral character of the active sources is known and controlled, there is no a priori knowledge or control over the noise sources that generate the passive data. As such, the weighting matrices above are also not known a priori. One method for determining the weighting matrices in a data-driven fashion is to optimize for the weights by minimizing the objective function: where (^^^^) is a target, user-chosen frequency spectrum corresponding to a desired frequency spectrum for the output pseudo-source data, e.g., a fixed-bandwidth flat spectrum, or Gaussian / Ricker-wavelet spectrum over a desired output frequency range. This approach to producing single pseudo-source data comprises the following steps: 1. Gather active and passive data (after initial noise analysis and data sorting) from each receiver into appropriate data set pairs; 2. Initialize weight matrices to any appropriate non-zero choice, e.g., unit weights; 3. Update the weight matrices for each pseudo-source receiver pair by optimizing the provided metric for matching a user-chosen pseudo- source spectrum; and 4. Recalculate final desired single pseudo-source responses from both passive and active data by applying the spectrum-optimized weight- matrices. Figures 4A and 4B illustrate some alternative configurations for the target area 2 (and consequently also the sub-surface target volume underneath this). As in a situation where two parallel cables are used more than one cable 6 is present in both cases, and each cable is positioned at or near the earth’s surface and is connected at one end to an interrogator 12 to collect signals from the cable. The cables also surround and delimit a target area 2 on the surface of the earth. The fact that the cables 6 contain, and in some cases also define, the target area means that surface waves travelling from one cable to another through the sub-surface target volume are detected. Source lines 8 are then positioned to roughly follow the shape of the perimeter of the target area 2, but are located outward of this perimeter in a horizontal direction. In the example shown in figure 4A, the target area 2 is shaped as a rectangle which narrows at one end. DAS cables 6 (three in this case) extend from a single interrogator 12 located at the narrow end of the target area 2. As an alternative to the setup shown, the central DAS cable can be dispensed with and only the two outermost cables can be present. However, the inclusion of the central cable would further aid in improving the final imaging quality by adding data samples of surface waves from within the target area, in additional to the boundary data (a minimum requirement). The source line extends to completely surround the target area in this case. The example shown in figure 4B uses two DAS cables 6 which are positioned to include a roughly 90 degree turn around halfway along their length in order to substantially surround a rectangular target area 2. The source line 8, again, mimics the shape of the target area 2 and completely surrounds it. The general design of the whole experiment is as follows. Firstly, the extent, depth and shape of the target area and volume are determined based on the area over which it is desired to infer subsurface geological properties. The resolution requirements, which are dictated by the smallest scales of geologic features that are sought after, are also determined. Once the target area and resolution are defined, the target imaging area (the sub-surface target volume) is defined to encompass the target geology area while ensuring that it is fully sampled in space by surface waves passing every point from many propagation directions. With these details defined, the specific layout of DAS cables and sources are selected accordingly. Only a single source line or very few source lines positioned to extend at least part of the way around the target area are required in all cases, including the parallel cable setup, rather than the traditional source carpet in which a grid of source activations is set up and sources need to be towed back and forth across a large area directly above a target zone.

Claims

Claims 1. A method for performing a seismic survey, comprising: positioning at least two fiber optic cables and defining a target area that is located in between the cables and delimits the top of a sub-surface target volume; coupling an end of each cable to a distributed acoustic sensing interrogator; and activating one or more seismic sources exclusively outside of the target area.

2. The method of claim 1, where activating the one or more seismic sources comprises activating the seismic sources along at least one source line, wherein the at least one source line is offset horizontally from the perimeter of the target area along the entire length of the source line, such that the at least one source line follows the shape of the target area perimeter and at least partially surrounds the target area.

3. The method of claim 2, wherein the at least one source line extends along or around the target area so that only a single source line is crossed when travelling in a direction outwards from the target area.

4. The method of any of claims 2 and 3, wherein the at least one source line is offset horizontally from the target area perimeter by between 500 meters and 2000 meters along the entire length of the source line.

5. The method of any of claims 2 to 4, wherein the at least one source line extends along at least two thirds of the length of the target area perimeter.

6. The method of claim 5, wherein the at least one source line completely surrounds the target area.

7. The method of claims 1 to 6, wherein activating the one or more seismic sources comprises activating the sources at locations which define a single open loop or a single closed loop.

8. The method of any of claims 1 to 7, comprising acquiring, by the interrogator coupled to that cable, data from each of the cables.

9. The method of claim 8, comprising processing the acquired data to produce a 3D representation of the sub-surface target volume.

10. The method of any of claims 8 and 9, wherein acquiring the data comprises acquiring passive data during a period in which the one or more seismic sources are not activated and acquiring active data simultaneously with activating the one or more seismic sources.

11. The method of claim 10, comprising, after acquiring the passive data and before acquiring the active data, determining signal bandwidth and / or signal-to-noise levels from the passive data and selecting one or more of the shot positions and spectral properties for the one or more seismic sources to be used during active data acquisition based on the determined signal bandwidth and / or signal-to-noise levels.

12. The method of any of claims 10 and 11, comprising: combining the active and passive data; producing a pseudo-source dataset from the combined data using interferometry methods; and producing a 3D representation of the sub-surface target volume from the pseudo-source data using surface wave inversion.

13. The method of any of claims 10 to 12, comprising selecting the frequencies emitted by the one or more seismic sources so that the active data contains a partial spectral overlap with the passive data and the frequencies emitted by the one or more seismic sources extend beyond those present in the passive data.

14. The method of any of claims 1 to 13, comprising activating the one or more seismic sources along at least one source line which extends along the perimeter of the target area on at least two opposing sides of the target area.

15. The method of any of claims 1 to 14, wherein coupling an end of each cable to a distributed acoustic sensing interrogator comprises coupling an end of each cable to a different distributed acoustic sensing interrogator.

16. An apparatus for carrying out a seismic survey, comprising: at least two DAS cables positioned outside of a target area that is located in between the cables and delimits the top of a sub-surface target volume; an interrogator coupled to an end of each cable; and at least one source line defined as a pathway along which one or more seismic sources for the survey are to be activated, wherein the at least one source line is exclusively outside of the target area.

17. The apparatus of claim 16, wherein the at least one source line is offset horizontally from the perimeter of the target area along the entire length of the source line, such that the at least one source line follows the shape of the target area perimeter and at least partially surrounds the target area.

Images