Conventional Seismic Data Research Articles

New insights into the Bedrock and Quaternary Geology of St. George’s Bay from a vertical integration of marine datasets, offshore western Newfoundland

St. George’s Bay lies south of Port au Port Peninsula on the west coast of Newfoundland, Canada. The Carboniferous Bay St. George subbasin extends beneath the bay and onshore to the southeast. The subbasin is an extension of the Magdalen Basin, a Carboniferous basin with probable resource potential, part of the broader Maritimes Basin. The onshore geology of the subbasin is relatively well understood. Marine datasets from the less well known offshore portion include high-resolution bathymetry, aeromagnetic data, shallow and industry wells, and both shallow and deeper (conventional) two-dimensional seismic data. Quaternary features were previously documented in inner St. George’s Bay where the surficial cover is thick. In the outer part of the bay, bedrock ridges are draped by iceberg-turbated glaciomarine mud and moraines. Beneath this cover, reinterpreted palynological analyses in the A-36 well reflect only Visean Codroy Group. The uppermost Codroy Group of this well may correlate with nearby red sandstones from shallow marine bedrock drill cores; however, younger Carboniferous rocks may also be present. Reflectors mapped using conventional seismic data mark the base Codroy Group and the top of a salt unit in the Codroy Group. Salt-cored anticlines and synclines are associated with halokinesis, and with a fault system with both dip-slip and dextral-slip components. Multibeam bathymetry and the second vertical derivative of the magnetic field delineate the folds as curvilinear, en echelon, and, in some cases, doubly plunging. Beneath the subbasin, seismic and aeromagnetic data indicate Proterozoic basement blocks and an offshore extension of the early Paleozoic Romaines Brook Fault. The Carboniferous structures suggest an environment of dextral transtension that reactivated early Paleozoic structures aligned with pre-existing weaknesses in basement.

Seismic reflection survey at Ayer Hangat site to investigate shallow subsurface structures

Ayer Hangat site is located in the island of Langkawi, northwest Malaysia. The site is characterized by the presence of hot spring. This hot spring is believed to be related to granitic intrusion nearby. Hence the present work is focusing on defining the shallow subsurface structures that control the migration of hot water to the surface. Seismic reflection method is used to achieve the goal of the present study. Forty three shot points were used with an offset of 5m of the nearest geophone. The shot-points interval is set to 1m. Seismograms were recorded on 24 channel TERRALOC instrument. The Geophone interval used was 1m. Conventional seismic data processing scheme was adopted. However, due to the fact that TERRALOC produce SEG2 data files, a script based on Obspy was written and used to convert to SEG-Y format. Afterwards, analyses were carried out using SU Package. The processed data is used to develop a model for the subsurface controlling structures. Such model will help in the understanding of the geothermal hot spring system in the area.

Delineating thin sand connectivity in a complex fluvial system in Mangala field, India, using high resolution seismic data

The Mangala field is located in the northern part of the onshore Barmer Basin in India. The primary reservoir in the field is the Fatehgarh Formation, deposited during the rifting phase that created the Barmer Basin during the Late Cretaceous to Early Paleocene period. The majority of reservoired oil is con¬tained within the Upper FM1 member of the Fatehgarh Formation, composed of single storey and multi-storey stacked, meandering channel sands. The average gross thickness of FM1 is 80 m, and individual sands vary in thickness from 3 to 7 m, with net-to-gross ratio ranging from 18% to 78% due to inherent heterogeneity within FM1 as evident from core data. For such a heterogeneous fluvial system, correlation of fluvial channel sands and flood plain shales poses a major challenge for reservoir characterization when based on well data alone. Detecting or mapping the lateral continuity of these thin fluvial channel sands is difficult because they are below seismic resolution in conventional seismic data. We applied sparse-layer reflectivity inversion (Zhang and Castagna, 2011) to the 3D stack PSTM data, which resulted in a data¬set with improved detectability and resolution. The 7-50 Hz bandwidth of the input seismic data increased to 7-100 Hz through the inversion process. Results were validated using well log and production data. The new data contributed to greater understanding of the lateral connectivity of the FM1 fluvial channel sands, and enabled geobody extraction for reservoir static modelling.

Provenance of the Lower Triassic Bunter Sandstone Formation: implications for distribution and architecture of aeolian vs. fluvial reservoirs in the North German Basin

AbstractZircon U–Pb geochronometry, heavy mineral analyses and conventional seismic reflection data were used to interpret the provenance of the Lower Triassic Bunter Sandstone Formation. The succession was sampled in five Danish wells in the northern part of the North German Basin. The results show that sediment supply was mainly derived from the Ringkøbing‐Fyn High situated north of the basin and from the Variscan belt located south of the basin. Seismic reflection data document that the Ringkøbing‐Fyn High was a local barrier for sediment transport during the Early Triassic. Hence, the Fennoscandian Shield did not supply much sediment to the basin as opposed to what was previously believed. Sediment from the Variscan belt was transported by wind activity across the North German Basin when it was dried out during deposition of the aeolian part of the Volpriehausen Member (lower Bunter Sandstone). Fluvial sand was supplied from the Ringkøbing‐Fyn High to the basin during precipitation events which occurred most frequently when the Solling Member was deposited (upper Bunter Sandstone). Late Neoproterozoic to Carboniferous zircon ages predominate in the Volpriehausen Member where the dominant age population with a peak age of 337 Ma corresponds to the culmination of Variscan high‐grade metamorphism, whereas a secondary age population with a peak at 300 Ma matches the timing of volcanism and magmatism at the Carboniferous/Permian boundary in the northern Variscan belt. Parts of the basement in the Ringkøbing‐Fyn High were outcropping during the Early Triassic and zircon ages similar to this Mesoproterozoic basement are present in the Bunter Sandstone. The heavy mineral assemblage of the Solling Member is uniform and has a high garnet content compared to the contemporaneous sediments in the Norwegian‐Danish Basin and in the southern part of the North German Basin. This finding confirms that a local source in the Ringkøbing‐Fyn High supplied most of the fluvial sediment in the northern part of the North German Basin. The northernmost part of the Bunter Sandstone is situated on a platform area that is separated from the basin area by a broad WNW–ESE‐oriented fault zone. The most promising reservoir in the basin area is the aeolian Volpriehausen Member since the sandstone has a wide lateral distribution and a constant thickness. The alluvial to ephemeral fluvial Solling Member may be a good reservoir in the platform area and marginal basin area, but the complex sand‐body architecture makes it difficult to predict the reservoir quality.

The application of quantitative gas saturation estimation based on the seismic wave dispersion inversion

Trace volumes of pore gas result in a drastic reduction of P-wave velocity, making it hard to determine the degree of gas saturation in a reservoir from P-wave velocity or related seismic attributes. An analysis of the seismic data suggests that the presence of hydrocarbons in reservoir units results in higher relative degrees of seismic wave dispersion and attenuation. These effects are generally ignored during a conventional seismic data analysis and inversion. In this paper, we applied a crossplot inversion method to estimate gas saturation. Based on the modeling results, a useful method for the quantitative determination of gas saturation is established based on the poststack seismic dataset. Gas saturation is a frequency-dependent attribute, which allows a crossplot inversion by way of a time–frequency decomposition of the seismic data. The modeling of reflections from the interface between a medium that disperses seismic waves and its elastic overburden indicates that the reflection coefficient is frequency-dependent and varies with gas saturation. This relationship may in turn vary with the numerical model used to describe the reservoir medium (e.g., sand) and with the depositional environment, which affect the amplitude-versus-offset behavior at the interface. Applying this method to field seismic data shows frequency-dependent anomalies similar to those predicted by the model. The gas saturation predicted by these methods has also been verified by well data.

Varying-Window-Length TFPF in High-Resolution Radon Domain for Seismic Random Noise Attenuation

The time–frequency peak filtering (TFPF) algorithm is an effective method for seismic random noise attenuation. The conventional TFPF filters seismic data only along the channel direction, ignoring the spatial characteristics of the reflection events, which results in the loss of directional information. In order to improve the filtering performance of TFPF, we adopt a Radon transform to implement spatiotemporal 2-D TFPF, which is doing TFPF in the Radon domain. Since this method takes the spatial correlation of the reflection events into account, it could extract the reflection events better than the conventional TFPF. As the conventional Radon transform may produce a smearing phenomenon, we apply an improved least squares high-resolution Radon transform to assist TFPF in this letter. Thus, the events can be highly focused to energy points in different positions of the Radon domain. Then we identify the signal parts and process them by TFPF with short window length (WL). The other parts are considered as noise, and we process them by TFPF with long WL. By virtue of this new method, we can preserve the valid signal better and suppress the random noise more effectively. Through experiments on the synthetic seismic records and the field seismic data, the new method possesses a superior performance in random noise attenuation and seismic event preservation compared with the conventional TFPF and Radon TFPF methods.

Increasing the effective bandwidth of seismic data through sparse-layer inversion: A case study from Mangala field, Rajasthan, India

Mangala field is in the northern part of the onshore Barmer Basin in India. The primary reservoir in the field is the Fatehgarh Formation, deposited in the rifting phase that created the Barmer Basin during the Late Cretaceous to early Paleocene period. The majority of reservoir oil is contained within the upper FM1 member of the Fatehgarh Formation, composed of single-story and multistory stacked, meandering-channel sands. These sands vary in thickness from 3 to 7 m, with net to gross ranging from 18% to 78%. For such a heterogeneous fluvial system, correlation of floodplain shales and fluvial sands poses a major challenge for reservoir characterization when based on well data alone. Conventional 3D seismic data do not resolve the thin FM1 sands. For more accurate reservoir modeling, various techniques, which included sparse-spike inversion and spectral decomposition, were attempted with limited success. Sparse-layer reflectivity inversion performed on 3D stack PSTM seismic data, however, resulted in improved detectability and resolution and has provided a greater understanding of the lateral continuity of these thin fluvial reservoir units. The 7- to 50-Hz bandwidth of the input seismic data increased to 7 to 100 Hz by the inversion process. The improved imaging of channel geometries has enabled geobody mapping and their input into a revised geologic model.

Migration of surface correlated multiples based on one-way wave equation

In conventional seismic data processing, multiple reflection waves are regarded as noise which should be removed from seismic data in order to avoid misunderstanding. In fact, multiples are also signals which contribute to part of subsurface structure illumination and imaging. In this paper, we analyzed the artifact generating mechanism and phenomenon caused by multiples in conventional one-way equation based pre-stack depth migration. In order to realize multiple imaging based on one-way wave equation, the boundary condition of one-way wave-equation based pre-stack depth migration was modified. It means that the input source wave field and recorded wave field are replaced by the record containing multiples and predicted surface correlated multiples, respectively. Then we stacked the pre-stack depth migration results from primaries and multiples based on the matching operator which is obtained from the norm II least-square energy difference methods. Imaging with multiples is a good compensation to imaging to that with primaries. The test on a simple model shows the validity of our method. Finally, jointed imaging with primaries and multiples was performed on the Sigsbee2B model. The results show that the imaging of the steep salt boundary was improved dramatically.

Signal-to-noise ratio application to seismic marker analysis and fracture detection

Seismic data with high signal-to-noise ratios (SNRs) are useful in reservoir exploration. To obtain high SNR seismic data, significant effort is required to achieve noise attenuation in seismic data processing, which is costly in materials, and human and financial resources. We introduce a method for improving the SNR of seismic data. The SNR is calculated by using the frequency domain method. Furthermore, we optimize and discuss the critical parameters and calculation procedure. We applied the proposed method on real data and found that the SNR is high in the seismic marker and low in the fracture zone. Consequently, this can be used to extract detailed information about fracture zones that are inferred by structural analysis but not observed in conventional seismic data.

Fining up the seismic stochastically for reservoir characterization, offshore North West Australia

SUMMARY The vertical resolution of conventional seismic data limits the illumination and definition of thin sand reservoirs offshore North West Australia. With the availability of more wells within a field it is justified to use a well-centric stochastic seismic inversion approach to characterize the reservoirs. The stochastic AVO seismic inversion yields multiple triplet realizations of acoustic impedance, Poisson’s ratio and density at the fine enough vertical sampling for a more relevant reservoir modelling procedure. The paper will discuss the stochastic seismic inversion approach, followed by a structured analysis workflow which condenses the generated multiple realizations to more interpretable data volumes.

A Geosteering Success Story From the North Sea

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 159132, ’Geosteering the Impossible Well: A Success Story From the North Sea,’ by Soazig Leveque and Christophe Dupuis, Schlumberger; Thomas Stærmose, Estela Vazquez Esmerode, and Peter Sommer Linnet, DONG E&P; and Graham Raeper, Schlumberger, prepared for the 2012 SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, 8-10 October. The paper has not been peer reviewed. The Nini East field represent a significant drilling challenge. The target is a post-depositionally remobilized sand reservoir with a thickness of 2‒15 m. None of the standard well-placement methods or current mapping tools was a suitable solution for the well objectives. A field test of the next-generation bed-boundary mapping tool was launched to geosteer two upcoming production wells. The two producers achieved an outstanding 0.99 and 0.96 net pay ratio with no sidetracks or delays and were completed within budget. Introduction The Nini East field is located 3 km south of the border between the Danish and Norwegian sectors of the North Sea (Fig. 1) and was discovered in 2001, when oil was found in the Kolga-member sandstones of the Upper Paleocene/Lower Eocene Sele formation. The challenges in drilling horizontal wells in this type of reservoir include a complex geological setting—leading, in some wells, to a low net-pay ratio—combined with the unstable reservoir- capping Sele shale, which increases the risk of having to drill sidetracks. The Kolga-member reservoir of the Nini East field consists of a glauconite-rich sandstone with unique characteristics. Although initially deposited as traditional gravity-flow sands, the Kolga sands have been remobilized under burial, leading to sand fluidization and injection along weakness zones. These traits create difficulty in defining the exact stratigraphic location of the sand. With these challenges in mind, the Dansk Olie og Naturgas company (DONG E&P) launched the development of the Nini East field in 2009 with a drilling campaign including two horizontal producers and one injector in the Kolga sand. The two main goals were to increase the percentage of horizontal section in pay while avoiding the Sele shale and maintaining the oil/water-contact (OWC) standoff. Methodology The operator decided that much-needed uncertainty reduction could only come from significant investment in seismic-data processing and inversion, combined with the new- generation deep-directional-electromagnetic (DDE) service-placement workflow. The seismic data were reprocessed and inverted to improve the imaging of the injectite sands. The seismic resolution was improved from approximately 10 m on the conventional seismic data to approximately 5 m on the inverted cube (Fig. 2). The primary method of trajectory adjustment involves correlating the logging-while-drilling (LWD) log response acquired on a drilling well with offset data acquired along previously drilled wells in the area. This implicitly relies on the assumption that the formation is continuous between the offset wells and the current drilling location—an assumption that may not hold in faulted reservoirs, pinchouts, unconformities, or highly remobilized reservoirs such as those found in the Nini East field.

Seismic prediction of sandstone diagenetic facies: Applied to Cretaceous Qingshankou Formation in Qijia Depression, Songliao Basin, East China

Abstract Seismic diagenetic facies is an important control on reservoir quality. This study investigated the feasibility of predicting sandstone diagenetic facies using conventional 3D seismic data by seismic sedimentology and calibrated by laboratory core-analysis data in the Qijia area of the Songliao Basin. Three core issues are related to seismic characterization of diagenetic facies, including how to correlate stratigraphic and diagenetic units from seismic data, how to evaluate the relationship between core-based diagenetic facies and seismic attributes, and how to find an effective way of mapping seismic diagenetic facies. Well- and seismic-based, high-resolution sequence analysis and seismic stratal slicing provide reservoir-scale (20 m) seismic representations of diagenetic units. Core-based analyses of sandstone diagenetic processes and diagenetic sequences reveal the kind of diagenesis that most influences reservoir quality. An investigation of reservoir parameters and acoustic rock properties further reveals the link between diagenetic facies and impedance, leading to a recognition of calcite cementation as the process that can be detected by seismic data. A seismic-based lithology cube (e.g., 90°-phased seismic volume) provides the amplitude (impedance) signal for detection of diagenetic facies. Stratal slices made from the seismic-based lithology cube are then used to interpret depositional facies and systems. Eventually a seismic diagenetic-facies map is generated through analysis of the relationships between depositional facies, impedance, and diagenetic facies. A case study of clay- and calcite-cemented sandstone in the Qijia area shows that although still in its infancy, seismic detection of sandstone diagenetic facies using conventional seismic data is definitely feasible.

Quaternary shelf structures SE of the South Island, imaged by high-resolution seismic profiling

Along the south-eastern coast of New Zealand's South Island, observations and characterisations of shelf geology are complicated by numerous possibly active faults (e.g. the coast-parallel Akatore and coast-perpendicular Waihemo and Castle Hill faults), a Miocene-aged volcanic edifice (i.e. the Dunedin volcano) and incision from an extensive submarine canyon system. Conventional marine seismic data do not adequately image the basin beneath the shallow shelf here. However, six recently digitised high-frequency single-channel boomer seismic surveys have enabled the investigation of unique local geological structures and their relationships to the tectonic and sedimentary development of the region. These structures have significant control on active processes such as: (1) the localisation of sedimentation and submarine erosion; (2) the instigation of canyon channel incision; and (3) the distribution of fluid migration pathways on the shallow shelf. Future data acquisition will further constrain these processes and help to evaluate earthquake risk in this region.

Tutorial: the seismic response to strong vertical velocity change

Conventional seismic data processing, whether pre-stack data conditioning or migration, is designed with the theory of P-wave reflected energy in mind for travel paths involving only a single reflection. Hence, any energy propagating with other modes or travel paths is not dealt with appropriately during conventional processing. It is primarily for this reason that we spend so much time pre-conditioning seismic data to meet the assumptions of the subsequent processes such as migration. Here I consider a typical North Sea environment where there are both strong vertical compaction gradients and high velocity contrasts in the layered Chalk Formation. I assess the behaviour of refracted and mode-converted seismic arrivals when subjected to conventional processing in order to explain how they give rise to anomalous events that contaminate migrated common reflection point gathers and images.

Broadband seismology in oil and gas exploration and production — Introduction

The last decade has seen considerable success toward solving long-standing exploration and reservoir characterization challenges, such as subsalt reservoir imaging. These advances have been enabled through new acquisition techniques in combination with advances in methods for processing, imaging and inversion of seismic data that is rich in low frequencies. In particular, methods such as reverse-time migration rely on highly accurate velocity models. Deriving such velocity models in turn fundamentally requires the acquisition of low-frequency seismic data. In addition, as is nicely demonstrated in one of the papers in this special section, low frequencies also benefit higher resolution images and enables deeper penetration in the subsurface (ten Kroode et al., 2013). The need for extending the temporal bandwidth of seismic data has therefore become increasingly evident and is a key area for research and development both in industry and academia. In this special section we have included papers both describing marine and land applications. We have included papers that The techniques for providing broadband frequency data in turn follow three different directions. First, seeking to widen the data bandwidth by filling in the notches caused by ghost arrivals, several papers in the special section present techniques for deghosting conventional marine seismic data. Second, aiming to push data acquisition toward …

Research and Application of Spectral Decomposition in Carbonate Reservoir and Fault Identification in MaiJie Let Gas Field, Amu Darya Basin

Spectral decomposition is to convert seismic signals from the time domain to the frequency domain by mathematical transformation, and analyze amplitude and phase response characteristics of different scale geological bodies. Spectral decomposition could get higher resolution than conventional seismic data. In the identification of the fault system, it is fit for fault interpretation and plane combination of the sections. In the reservoir prediction, it can identify the shape and contour of the reservoir. This document analyzed algorithm and applicability of short time Fourier transform, continuous wavelet transform and S-transform. Using these three methods for carbonate reservoir identification in MaiJie let gas field, Amu Darya Basin, it proves that the frequency division section is more clearly than conventional seismic section in reservoir and fault description. And S-transform gets the best result.

Variable-depth streamer acquisition: broadband data for imaging, post and pre-stack inversion

Variable-depth streamer acquisition is a broadband streamer acquisition technique where the depth profile of the streamer is optimized in order to ensure receiver ghost diversity, which in turn allows the deconvolution of the residual ghost at the imaging stage, pre-stack or post-stack. This technique allows the streamers to be towed at an average depth of several tens of meters, which combined with the use of solid streamers, ensures the raw data has an exceptionally good signal-to-noise ratio, especially at low frequencies. This variable-depth streamer acquisition and processing has been field-tested on a variety of locations, achieving bandwidth up to 6 octaves (2.5Hz -160 Hz). This broad bandwidth translates into improved results for the acoustic impedance inversion. The lack of low frequencies in conventional seismic data means that a low frequency model must be incorporated in the inversion process, obtained by interpolating low-passed filtered impedance logs between well locations. With variable-depth streamer data, high-resolution NMO-derived seismic velocities are used to define the low frequency model in a range 0-5Hz, while the reflectivity provides information from 2.5Hz. Variable-depth streamer data thus have the potential to fill the usual gap between the high frequencies of the seismic velocities and the low frequencies of the reflectivity, the 2.5-5 Hz octave being the overlapping zone. Pre-stack elastic inversion has also been performed, providing both impedance and Vp/Vs sections, proving the feasibility of pre-stack deghosting of variable-depth streamer data.

Mapping the base of sand dunes using a new design of land-streamer for static correction applications

The complex near-surface structure is a major problem in land seismic data. This is more critical when data acquisition takes place over sand dune surfaces, where the base of the sand acts as a trap for energy and, depending on its shape, can considerably distort conventionally acquired seismic data. Estimating the base of the sand dune surface can help model the sand dune and reduce its harmful influence on conventional seismic data. Among the current methods to do so are drilling upholes and using conventional seismic data to apply static correction. Both methods have costs and limitations. For upholes, the cost factor and their inability to provide a continuous model is well realized. Meanwhile, conventional seismic data lack the resolution necessary to obtain accurate modeling of the sand basement. We developed a method to estimate the sand base from land-streamer seismic acquisition that is developed and geared to sand surfaces. Seismic data acquisition took place over a sand surface in the Al-Thumamah area, where an uphole is located, using the developed land-streamer and conventional spiked geophone systems. Land-streamer acquisition not only provides a more efficient data acquisition system than the conventional spiked geophone approach, but also in our case, the land-streamer provided better quality data with a broader frequency bandwidth. Such data enabled us to do accurate near-surface velocity estimation that resulted in velocities that are very close to those measured using uphole methods. This fact is demonstrated on multiple lines acquired near upholes, and agreement between the seismic velocities and the upholes is high. The stacked depth seismic section shows three layers. The interface between the first and second layers is located at 7 m depth, while the interface between second and third layers is located at 68 m depth, which agrees with the uphole result.

Seismic imaging of the water-column deep layer associated with the Deepwater Horizon oil spill

The submerged hydrocarbon layer at a depth of approximately 1100–1200 m in the water column near the Deepwater Horizon well site was associated with a temperature anomaly that was imaged by seismic reflection data collected two days before and one day after the well shut-in. Conventional multichannel seismic reflection profiling data were processed to enhance water-column reflections by filtering out low-frequency energy and applying higher gain than typically used for imaging subsurface geology. The deep layer reflectivity, within a few km of the well site, has an impedance less than seawater by about a factor of approximately [Formula: see text], or equivalently, a negative sound-velocity step of about [Formula: see text]. Water-column profile data collected at the same time as the seismic data show a comparable sound velocity and temperature decrease associated with the deep layer. Mixing of less dense hydrocarbons entrains cold seawater, resulting in a temperature anomaly within the deep layer at the level of neutral buoyancy. This temperature contrast explains the negative sound-velocity anomaly associated with the deep reflector. Because of the small hydrocarbon concentration in the deep layer, their contribution to the seismic velocity anomaly is too small to be significant. The high horizontal resolution (3 m) of the seismic image reveals a layer with patchiness of several hundred meters lateral extent.

Improved Hilbert-Huang Transform and its Application in Seismic Data Spectrum Decomposition

In the conventional seismic data time-frequency analysis, the wavelet transform, wigner ville distribution and so on, cannot meet the high precision time-frequency analysis requirements because of uncertainty principle and cross-term interference. The recently popular Hilbert-Huang transform (HHT) although overcomes these conventional methods’ deficiencies; it still has some unsolved deficiencies due to the theory imperfect. This paper focuses on an improved HHT so as to ameliorate the defect of original HHT. First of all, the wavelet packet transform (WPT) as the preprocessing will be used to the inspected signal, to get some narrow band signals. Then use the empirical mode decomposition (EMD) on the narrow band signals and get the real intrinsic mode function (IMF) by the method of correlation coefficient. From the numerical study and comparison of improved HHT, wavelet transform and HHT, it proves the validity and advantages of this improved method. At last, the improved HHT is applied to marine seismic data by the spectrum decomposition technology, and it well reveals the low frequency shadow phenomenon of the reservoir. The results show that this new method has effectiveness and feasibility in seismic data spectrum decomposition.