Investigating electrical anisotropy in the Barents Sea

Energy companies have been exploring in the Barents Sea for more than 30 years. The Barents Sea has garnered significant attention recently with the high-profile Skrugard and Havis discoveries, the record interest in 72 of 86 blocks in the recent 22nd Norwegian licensing round, and the ongoing development of Snohvit and Goliat. These discoveries and developments demonstrate the importance and future potential of the Barents Sea as a major hydrocarbon-producing province.

Regular license rounds and favorable financial terms encouraging exploration have led geophysical contractors to acquire large multiclient datasets in the Barents Sea. The key to linking geophysical responses seen in these data and the underlying rock and fluid properties is a robust understanding of the rock physics within reservoir intervals. Rock Solid Images (RSI) has constructed a regional rock physics database for the Barents Sea comprising 79 key wells (Figure 1). This database has been licensed by many companies exploring in the region. The focus is to provide an atlas of seismic responses to allow geologists and geophysicists to understand how the seismic signature of a particular reservoir varies with changing reservoir quality and fluid content.

FIGURE 1. This map shows well locations and CSEM surveys in the Barents Sea. (CSEM survey information courtesy of EMGS; images courtesy of RSI)

In the last five years several large multi-client 3-D controlled-source electromagnetic (CSEM) surveys have been added to the data available to Barents Sea explorers. CSEM is a relatively new tool in the exploration toolbox and in the correct circumstances can provide higher sensitivity to hydrocarbon saturation than is possible to achieve with conventional seismic reflection data. For the true value of these data to be realized, the interpretation must be placed within a regional resistivity context, calibrated where possible to well log data.

The Barents Sea is geologically complex – stratigraphically, structurally, and historically. One component of this complexity is the presence of strong anisotropy in measured and derived electrical resistivity. RSI is undertaking a comprehensive two-year industry-funded study of electrical anisotropy in the Barents Sea using a combination of well log data, CSEM survey data, and rock physics modeling techniques. The overall objective of the study is to understand the geological controls on electrical anisotropy and build new predictive rock physics models that will allow more accurate predictions of rock and fluid properties of the Barents Sea reservoirs. To do this requires analysis of regional resistivity trends and investigation of their underlying cause(s). It is particularly important to develop an understanding of the causes of, and trends in, electrical anisotropy that can be significant in sedimentary structures. Disregarding resistivity anisotropy will lead to misleading CSEM survey feasibility studies, inaccurate CSEM data analysis, inaccurate estimations of hydrocarbon saturations, and, consequently, erroneous interpretations.

Method

To achieve the goals of this study, a three-stage workflow has been developed. The first stage is examining resistivity trends in major stratigraphic units across the area using standard induction well log data and looking at how these regional trends relate to age, compaction, lithology (especially clay content), and hydrocarbon saturation. Although these provide trends for the most part only in horizontal resistivity, they give useful guidelines on the variability likely to be encountered. Cross-plots of resistivity against seismic attributes such as impedance and Poisson’s ratio for both in situ and varying fluid saturations allow comparison of the sensitivity of CSEM and seismic-based attributes, providing an indication of the best attribute (or combination of attributes) for each situation.

Particular challenges for this stage include:

Porosity calculation. Porosity is calculated from the final edited bulk density using calculated mineral volumes and properties as well as fluid volumes and properties. The consortium is using core data to quality-check these measurements if available, although occasionally core values are higher than calculated total porosity. This is most likely due to laboratory-based helium methods in which the helium gas reads the microporosity, whereas the logs are not sensitive to this.

Water resistivity and saturation estimation. Water resistivity (R) and salinity are critical for understanding electrical resistivity, not only within reservoir units but also in the overburden structure. These are generally determined from a Pickett plot or from drillstem test data where available, although drilling fluid contamination often renders drillstem test results unreliable. Since in most cases the salinity data from drillstem test/produced water are unavailable, the Rvalue is estimated using the Pickett plot method. Water saturation (S) is calculated from either the Archie or Simandoux method, depending on the clay content of the reservoir in question.

Mud invasion correction. Occasionally water-based drilling mud invades hydrocarbon-bearing formations. This is clearly seen on resistivity profiles, where the shallow resistivity is reading very low (the drilling mud) and the deep is assumed to be reading the true formation resistivity. Depending on the degree of invasion, typically only the density will be edited assuming the deeper reading velocity logs are reading the true formation fluids. A straightforward fluid substitution is then performed iteratively until the true formation value is arrived at, which in hydrocarbon reservoirs is lower than the measured value.

FIGURE 2. In this example of a resistivity fluid substitution log plot for three wells (left to right), measured depth, gamma ray, water, hydrocarbon saturations, and modeled resistivity curves are shown per well. Resistivity curves show measured in situ resistivity (black), 20% hydrocarbon (purple), 100% wet (blue), 80% hydrocarbon (red), and in situ saturation (green).

The second stage investigates electrical anisotropy. Information on electrical anisotropy can be obtained from three sources:

• Three-component logging tools. Resistivity data from such tools provides a direct measurement of electrical anisotropy. However, such information is scarce and seldom available for the complete overburden section;
• Deviated well logs. Standard induction logs from deviated wells can be used to estimate electrical anisotropy in stratigraphic sections the wells penetrate. However, to achieve a robust estimate, several wells must penetrate the strata of interest at a range of angles. This condition is seldom satisfied except in mature fields; and
• CSEM data. Using multi-azimuth CSEM data, the bulk background electrical anisotropy in major stratigraphic units can be directly estimated using a structurally constrained modeling approach, with the horizontal resistivity tied to induction logs in the survey area. Where available, the anisotropy measure derived in this manner will be calibrated to the information from the first and second measurements to provide a more complete well tie. This approach is used where both CSEM and well log data are available. EMGS has made subsets of its extensive library of CSEM data in the Barents Sea available for this study.

This information is used to examine electrical anisotropy trends within major stratigraphic units across the region to build background models for CSEM sensitivity analysis. By performing fluid substitution in the main reservoir intervals, the 1-D CSEM sensitivity to variations in fluid saturation is calculated and compared at each of the wells included in the study. Although this does not give a complete analysis of CSEM sensitivity (which requires full 3-D analysis), it will provide a valuable indication of the classes of reservoir structure likely to be detected by CSEM surveys, the bounds on their properties required for CSEM sensitivity, and the variation in these across the region.

The results of the first two phases provide an empirical atlas of resistivity and anisotropy across the study region. This, combined with RSI’s underlying seismic attribute atlas, is a useful guide to physical properties and seismic/CSEM sensitivity to key parameters.

The goal of the third stage is to extend the analysis further by developing and calibrating electrical rock physics models and workflows that explain the observed trends. These rock physics models have two primary uses:

• They ultimately will allow electrical anisotropy to be predicted to within a quantified level of uncertainty from standard well log data. This will allow robust background models to be constructed, leading to a better understanding of CSEM sensitivity in new areas and more robust survey design when feasibility analysis demonstrates the potential value of CSEM acquisition and inversion; and
• For existing CSEM data these models will allow anisotropic resistivity information to be interpreted robustly in terms of the underlying rock and fluid properties of interest.

Observations to date

Analysis is still at a preliminary stage; however, results show high levels of electrical anisotropy. Once the data analysis is complete, understanding of the results will be enhanced through tight integration with the complex geological history of the Barents Sea, specifically that of uplift, tectonic inversion, and erosion. A future stage of the project will involve an analysis of seismic anisotropy vs. electrical anisotropy. For more information, visit rocksolidimages.com.