Despite the significant advances in seismic technology over the past decade or so, dry holes are still a much-too-common occurrence in the petroleum industry. There clearly is a compelling need for applications to detect the presence (or absence) of hydrocarbons in subsurface structures to help reduce drilling risk and increase exploration and production (E&P) success rates.

With a goal to fill this need, Spectraseis Inc. has concentrated on Direct Hydro-carbon Indicator (DHI) methodology since the company was founded in 2003. The organization and its network of research collaborators include globally recognized experts in numerical modeling of wave propagation in porous media.

The company’s extensive research and development efforts in DHI — including many field

Figure 1. Example of the frequency spectra for the vertical component of surface data yielding a rich set of spectral peaks. (Images courtesy of Spectraseis)
trials — evolved out of earlier research at the University of Zurich, which focused in large part on the discovery of intriguing amplitude peaks congregating around 3 Hz in surface seismic data measured above hydrocarbon reservoirs.

Historically, the seismic industry has ignored seismic data below 10 Hz, where conventional geophones are less effective. Yet these low frequencies now are recognized to harbor information which, when adequately interpreted, has the potential to significantly reduce the number of dry holes drilled.

As a result, passive seismic techniques using low-frequency ambient waves, which occur everywhere in the earth’s crust, are capturing considerable attention in the petroleum industry. These techniques were one of the major topics at a European Association of Geoscientists and Engineers (EAGE) workshop on passive seismic methods and applications held in Dubai in December 2006 that drew a crowd of 120 participants representing nearly all major E&P firms and contractors.

The company has considerable experience in data acquisition, data analysis and documentation of how low-frequency waves in the 1-Hz to 10-Hz range are shaped into coherent patterns above hydrocarbon-bearing reservoirs. The company has partnered with an array of national producers including Pemex, KOC and SRAK (a joint venture of Saudi Aramco, Shell and Repsol) and anticipates increased affiliations with international majors in 2007.

Central to the company’s work is the observation that the seismic background spectrum is modified differently via interactions with subsurface reservoirs having hydrocarbon-filled pores versus water-saturated formations. Hydrocarbon reservoirs produce characteristic patterns in the frequency spectra of surface signal, which can be analyzed to acquire information about the subsurface structures.

The ability to identify and map specific spectral signatures directly related to the presence of
Figure 2. Data from a survey in Brazil showing consistent anomalies in the Fourier spectra of surface velocities measured within and outside the boundaries of a known oil reservoir.
hydrocarbons is anticipated to have a dramatic impact on oil and gas industry exploration success rates.

Even though studies worldwide have confirmed that coherent patterns related to oil and gas-bearing reservoirs do, in fact, exist in the low-frequency domain, explanations for the underlying complex physical mechanisms of these events continue to challenge researchers. They include:
• Standing wave resonance: Occurs on a macro scale where characteristic maxima are generated owing to reflections between the reservoir and the surface — and within the reservoir — caused by complex impedance contrast due to the reservoir.
• Selective attenuation: Characteristic minima are caused by frequency-dependent attenuation within the reservoir. The company is working to determine the conditions whereby a selective, frequency-dependent attenuation could generate spectral anomalies coincident to the observed hydrocarbon micro-tremor signal.
• Resonant amplification effects of ambient seismic waves: Pore spaces that are partially hydrocarbon-filled demonstrate a resonance frequency. The organization is investigating conditions necessary to activate the resonance of the hydrocarbon-filled pores and the conditions that dictate whether the resonance frequency can be measured at the surface.
As the result of its efforts in the passive low-frequency seismic domain, the organization has developed a set of commercial tools for acquisition, processing and interpretation of low-frequency seismic data.

The company’s proprietary HyMAS (Hydrocarbon Micro-tremor Analysis) technology is used to identify and map the spectral signatures associated with suspected hydrocarbon-filled reservoirs. Hydrocarbon micro-tremors represent a frequency-dependent hydrocarbon signature present in incoming background waves. HyMAS analyzes the changes triggered in the background wave spectrum when the low-frequency waves propagate through hydrocarbon reservoirs. The technology can discern specific spectral signatures characteristic of the interaction between the reservoir and its fluid content, thereby answering a critical question not addressed by expensive, time-consuming 3-D seismic surveys, i.e., whether the structure of interest contains hydrocarbons.

Reliable information, fast turnaround, low cost and a minimal environmental footprint are hallmarks of the HyMAS technology, which is a relatively straightforward application.
The program utilizes highly sensitive, portable three-component broadband seismometers to acquire the complex low-frequency data. The instruments are placed in shallow holes about 1.5 ft (0.5 m) deep for protection from the elements and to improve coupling. The acquired data are filtered to separate artificial and surface-generated signals from the spectral patterns related to subsurface structures.

During the acquisition process, specialized survey teams move among a series of designated points covering a grid layout with node spacing of 820 ft to 6,560 ft (250 m to 2,000 m), depending on the survey goals. Using preliminary processing results, the survey layout can be adjusted on the fly in order to increase the resolution in areas of high interest.
Multiple teams can be deployed at the same time, and three teams can typically acquire 40 sq miles (100 sq km) of medium-resolution data in 30 days or less.

The acquired HyMAS raw data are subjected to the company’s RIO processing software suite in order to remove or attenuate all signals unrelated to subsurface structures, e.g., surface noise, and to correct the data set for diurnal and near surface geology-related variations.
A recently developed, patent-pending technique that applies an auto-normalization method to resolve signal variations over time is included in RIO’s more than 50 processing and analysis routines.

The low-frequency data can be interactively overlaid with geological and geophysical information such as contour maps and fault maps, and a variety of mathematical operators can be applied for multi-attribute cross correlation as well as visual interpretation.
DHI technology utilizing low-frequency passive seismic data is applicable for exploration, field appraisal and production and is especially valuable in areas having little subsurface information. Despite its potential to reduce the number of dry holes drilled, however, the technique is not a standalone tool – except in early exploration efforts – but a supplement to other technologies.

For example, it can be used to screen large areas to detect specific locales where
Figure 3. Frequency spectra resulting from proprietary Hymas processing which demonstrates a strong signal anomaly around 3 Hz associated with a known hydrocarbon reservoir.
hydrocarbon presence is indicated. This approach can save time and money by pinpointing specific areas of interest where the operator might then choose to apply 3-D seismic versus shooting expensive 3-D surveys over the entire region.

Stratigraphic traps, which can be difficult to map using either 2-D or 3-D seismic, are good targets for low-frequency passive seismic evaluation. Still another target area is the deep water, where the search continues for big finds and well costs are best described as astronomical.

As testimony to the efficacy of the technology, the first full-scale commercial survey using HyMAS was a recognized success.

The project took place in 2004 in the onshore Potiguar Basin in Brazil at the complex producing Mossoro field operated by Petrobras. One of the main catalysts for the survey was poor-quality 3-D seismic data that was difficult to interpret over large parts of the field.
The organization implemented the HyMAS effort as a “blind test” with no prior knowledge of the geology or the reservoir parameters. The survey area covered about 45 sq miles (120 sq km) and was completed in 55 days.

When the resulting HyMAS hydrocarbon potential maps were superimposed over the existing Petrobras-generated reservoir maps, HyMAS demonstrated a correlation of more than 50% for the uncalibrated maps and 85% for the calibrated maps.

The test project identified two producing zones quite clearly and partially revealed a third
Figure 4. Full Navier-Stokes modeling of fluid movement in the pore space done in cooperation with ASCOMP Ltd. demonstrating the potential of a low frequency resonance as an explanation for hydrocarbon-related icro-tremors.
one. Also, a strong positive correlation was observed between signal amplitudes and oil column thickness measured by eight logged wells. The discrepancy relating to the third producing zone was resolved in 2006 using newly developed processing techniques. Further surveys are underway in the same region.

The service company intends to substantially expand its existing trove of low-frequency data via nine new surveys in the first half of 2007, including a marine trial using recoverable ocean bottom sensors deployed over a proven non-producing field in the North Sea. The collaborative effort will include company shareholder Norsk Hydro and Scripps Institution of Oceanography.

Concurrently, the company is going full tilt investigating and refining new technology to add to its sizeable toolkit.

For example, a recently completed numerical feasibility study determined that time-reverse modeling is a promising technique for localizing hydrocarbon micro-tremors. The company demonstrated that the reverse modeling approach may be used not only to localize the origin of tremor signals but also to detect stacked reservoirs.